Description #

LLaVA multimodal image understanding model – visual semantic parsing and natural language description generation

Key features #

• Multimodal input processing
• Supports three image formats: URL, Base64 encoding, and binary stream
• Visual-language joint modeling
• Accurately maps image content to semantic space
• Conversational prompt support
• Guides generation direction via the prompt parameter
• Resource usage
• It is recommended to use a GPU with 48GB or more VRAM

Scenario optimization #

• Optimization for mixed Chinese and English scenarios: specifically enhances Chinese semantic understanding
• Supports correcting image size to be consistent; it is recommended to set image size based on image and GPU conditions

Operator parameters #

Input #

Output #

The processed array, with each element representing the visual understanding result for each image.

Parameters #

If a parameter does not have a default value, it is required

Examples #

The following code demonstrates how to use daft to run the operator for image content understanding and generate descriptions according to instructions.

from __future__ import annotations

import os

import daft
from daft import col
from daft.las.functions.multimodal.llava_one_vision_image_understanding import LlavaOneVisionImageUnderstanding
from daft.las.functions.udf import las_udf

if __name__ == "__main__":
    if os.getenv("DAFT_RUNNER", "native") == "ray":
        import logging

        import ray

        def configure_logging():
            logging.basicConfig(
                level=logging.INFO,
                format="%(asctime)s - %(name)s - %(levelname)s - %(message)s",
                datefmt="%Y-%m-%d %H:%M:%S.%s".format(),
            )
            logging.getLogger("tracing.span").setLevel(logging.WARNING)
            logging.getLogger("daft_io.stats").setLevel(logging.WARNING)
            logging.getLogger("DaftStatisticsManager").setLevel(logging.WARNING)
            logging.getLogger("DaftFlotillaScheduler").setLevel(logging.WARNING)
            logging.getLogger("DaftFlotillaDispatcher").setLevel(logging.WARNING)

        ray.init(dashboard_host="0.0.0.0", runtime_env={"worker_process_setup_hook": configure_logging})
        daft.set_runner_ray()
    daft.set_execution_config(actor_udf_ready_timeout=600)
    daft.set_execution_config(min_cpu_per_task=0)

    tos_dir_url = os.getenv("TOS_DIR_URL", "las-cn-beijing-public-online.tos-cn-beijing.volces.com")
    samples = {
        "image_path": [
            f"https://{tos_dir_url}/public/shared_image_dataset/cat_ip_adapter.jpeg"
        ],
        "prompt": [""]  
    }

    image_src_type = "image_url"
    model_path = os.getenv("MODEL_PATH", "/opt/las/models")
    model_name = os.getenv("MODEL_NAME", "LLaVA-OneVision-1.5-8B-Instruct")
    dtype = "bfloat16"
    use_flash_attention_2 = True
    prompt = None
    max_caption_length = 256
    resized_height = None
    resized_width = None
    batch_size = 2
    rank = 0

    ds = daft.from_pydict(samples)
    ds = ds.with_column(
        "caption",
        las_udf(
            LlavaOneVisionImageUnderstanding,
            construct_args={
                "image_src_type": image_src_type,
                "model_path": model_path,
                "model_name": model_name,
                "dtype": dtype,
                "use_flash_attention_2": use_flash_attention_2,
                "prompt": prompt,
                "max_caption_length": max_caption_length,
                "resized_height": resized_height,
                "resized_width": resized_width,
                "batch_size": batch_size,
                "rank": rank,
            },
            num_gpus=1,
            batch_size=2,
            concurrency=1,
        )(col("image_path"), col("prompt")),
    )

    ds.show()

    # ╭────────────────────────────────┬────────────────────┬──────────────────────────────────────────────────────────╮
    # │ image_path                     ┆ prompt             ┆ caption                                                  │
    # │ ---                            ┆ ---                ┆ ---                                                      │
    # │ Utf8                           ┆ Utf8               ┆ Utf8                                                     │
    # ╞════════════════════════════════╪════════════════════╪══════════════════════════════════════════════════════════╡
    # │ tos://las-cn-beijing-public-o… ┆ Please specify the type of the image. ┆ This image belongs to the animation category, specifically CGI (computer-generated imagery) animation…          │
    # ╰────────────────────────────────┴────────────────────┴──────────────────────────────────────────────────────────╯