mplug_owl2/serve/cli.py

import argparse
import torch

from mplug_owl2.constants import IMAGE_TOKEN_INDEX, DEFAULT_IMAGE_TOKEN
from mplug_owl2.conversation import conv_templates, SeparatorStyle
from mplug_owl2.model.builder import load_pretrained_model
from mplug_owl2.mm_utils import process_images, tokenizer_image_token, get_model_name_from_path, KeywordsStoppingCriteria

from PIL import Image

import requests
from PIL import Image
from io import BytesIO
from transformers import TextStreamer


def disable_torch_init():
    """
    Disable the redundant torch default initialization to accelerate model creation.
    """
    import torch
    setattr(torch.nn.Linear, "reset_parameters", lambda self: None)
    setattr(torch.nn.LayerNorm, "reset_parameters", lambda self: None)


def load_image(image_file):
    if image_file.startswith('http://') or image_file.startswith('https://'):
        response = requests.get(image_file)
        image = Image.open(BytesIO(response.content)).convert('RGB')
    else:
        image = Image.open(image_file).convert('RGB')
    return image


def main(args):
    # Model
    disable_torch_init()

    model_name = get_model_name_from_path(args.model_path)
    tokenizer, model, image_processor, context_len = load_pretrained_model(args.model_path, args.model_base, model_name, args.load_8bit, args.load_4bit, device=args.device)

    conv_mode = "mplug_owl2"

    if args.conv_mode is not None and conv_mode != args.conv_mode:
        print('[WARNING] the auto inferred conversation mode is {}, while `--conv-mode` is {}, using {}'.format(conv_mode, args.conv_mode, args.conv_mode))
    else:
        args.conv_mode = conv_mode

    conv = conv_templates[args.conv_mode].copy()
    roles = conv.roles

    image = load_image(args.image_file)
    # Similar operation in model_worker.py
    image_tensor = process_images([image], image_processor, args)
    if type(image_tensor) is list:
        image_tensor = [image.to(model.device, dtype=torch.float16) for image in image_tensor]
    else:
        image_tensor = image_tensor.to(model.device, dtype=torch.float16)

    while True:
        try:
            inp = input(f"{roles[0]}: ")
        except EOFError:
            inp = ""
        if not inp:
            print("exit...")
            break

        print(f"{roles[1]}: ", end="")

        if image is not None:
            # first message
            inp = DEFAULT_IMAGE_TOKEN + inp
            conv.append_message(conv.roles[0], inp)
            image = None
        else:
            # later messages
            conv.append_message(conv.roles[0], inp)
        conv.append_message(conv.roles[1], None)
        prompt = conv.get_prompt()

        input_ids = tokenizer_image_token(prompt, tokenizer, IMAGE_TOKEN_INDEX, return_tensors='pt').unsqueeze(0).to(model.device)
        stop_str = conv.sep if conv.sep_style not in [SeparatorStyle.TWO, SeparatorStyle.TWO_NO_SYS] else conv.sep2
        keywords = [stop_str]
        stopping_criteria = KeywordsStoppingCriteria(keywords, tokenizer, input_ids)
        streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)

        with torch.inference_mode():
            output_ids = model.generate(
                input_ids,
                images=image_tensor,
                do_sample=True,
                temperature=args.temperature,
                max_new_tokens=args.max_new_tokens,
                streamer=streamer,
                use_cache=True,
                stopping_criteria=[stopping_criteria])

        outputs = tokenizer.decode(output_ids[0, input_ids.shape[1]:]).strip()
        conv.messages[-1][-1] = outputs

        if args.debug:
            print("\n", {"prompt": prompt, "outputs": outputs}, "\n")


if __name__ == "__main__":
    parser = argparse.ArgumentParser()
    parser.add_argument("--model-path", type=str, default="facebook/opt-350m")
    parser.add_argument("--model-base", type=str, default=None)
    parser.add_argument("--image-file", type=str, required=True)
    parser.add_argument("--device", type=str, default="cuda")
    parser.add_argument("--conv-mode", type=str, default=None)
    parser.add_argument("--temperature", type=float, default=0.2)
    parser.add_argument("--max-new-tokens", type=int, default=512)
    parser.add_argument("--load-8bit", action="store_true")
    parser.add_argument("--load-4bit", action="store_true")
    parser.add_argument("--debug", action="store_true")
    parser.add_argument("--image-aspect-ratio", type=str, default='pad')
    args = parser.parse_args()
    main(args)