mirror of
https://github.com/DS4SD/docling.git
synced 2025-12-16 16:48:21 +00:00
feat: [Experimental] Introduce VLM pipeline using HF AutoModelForVision2Seq, featuring SmolDocling model (#1054)
* Skeleton for SmolDocling model and VLM Pipeline Signed-off-by: Christoph Auer <cau@zurich.ibm.com> Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * wip smolDocling inference and vlm pipeline Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * WIP, first working code for inference of SmolDocling, and vlm pipeline assembly code, example included. Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Fixes to preserve page image and demo export to html Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Enabled figure support in vlm_pipeline Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Fix for table span compute in vlm_pipeline Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Properly propagating image data per page, together with predicted tags in VLM pipeline. This enables correct figure extraction and page numbers in provenances Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Cleaned up logs, added pages to vlm_pipeline, basic timing per page measurement in smol_docling models Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Replaced hardcoded otsl tokens with the ones from docling-core tokens.py enum Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Added tokens/sec measurement, improved example Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Added capability for vlm_pipeline to grab text from preconfigured backend Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Exposed "force_backend_text" as pipeline parameter Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Flipped keep_backend to True for vlm_pipeline assembly to work Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Updated vlm pipeline assembly and smol docling model code to support updated doctags Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Fixing doctags starting tag, that broke elements on first line during assembly Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Introduced SmolDoclingOptions to configure model parameters (such as query and artifacts path) via client code, see example in minimal_smol_docling. Provisioning for other potential vlm all-in-one models. Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Moved artifacts_path for SmolDocling into vlm_options instead of global pipeline option Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * New assembly code for latest model revision, updated prompt and parsing of doctags, updated logging Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Updated example of Smol Docling usage Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Added captions for the images for SmolDocling assembly code, improved provenance definition for all elements Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Update minimal smoldocling example Signed-off-by: Christoph Auer <cau@zurich.ibm.com> * Fix repo id Signed-off-by: Christoph Auer <cau@zurich.ibm.com> * Cleaned up unnecessary logging Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * More elegant solution in removing the input prompt Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * removed minimal_smol_docling example from CI checks Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Removed special html code wrapping when exporting to docling document, cleaned up comments Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Addressing PR comments, added enabled property to SmolDocling, and related VLM pipeline option, few other minor things Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Moved keep_backend = True to vlm pipeline Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * removed pipeline_options.generate_table_images from vlm_pipeline (deprecated in the pipelines) Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Added example on how to get original predicted doctags in minimal_smol_docling Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * removing changes from base_pipeline Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Replaced remaining strings to appropriate enums Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Updated poetry.lock Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * re-built poetry.lock Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> * Generalize and refactor VLM pipeline and models Signed-off-by: Christoph Auer <cau@zurich.ibm.com> * Rename example Signed-off-by: Christoph Auer <cau@zurich.ibm.com> * Move imports Signed-off-by: Christoph Auer <cau@zurich.ibm.com> * Expose control over using flash_attention_2 Signed-off-by: Christoph Auer <cau@zurich.ibm.com> * Fix VLM example exclusion in CI Signed-off-by: Christoph Auer <cau@zurich.ibm.com> * Add back device_map and accelerate Signed-off-by: Christoph Auer <cau@zurich.ibm.com> * Make drawing code resilient against bad bboxes Signed-off-by: Christoph Auer <cau@zurich.ibm.com> * chore: clean up code and comments Signed-off-by: Christoph Auer <cau@zurich.ibm.com> * chore: more cleanup Signed-off-by: Christoph Auer <cau@zurich.ibm.com> * chore: fix leftover .to(device) Signed-off-by: Christoph Auer <cau@zurich.ibm.com> * fix: add proper table provenance Signed-off-by: Christoph Auer <cau@zurich.ibm.com> --------- Signed-off-by: Christoph Auer <cau@zurich.ibm.com> Signed-off-by: Maksym Lysak <mly@zurich.ibm.com> Co-authored-by: Maksym Lysak <mly@zurich.ibm.com>
This commit is contained in:
Christoph Auer
GitHub
180
docling/models/hf_vlm_model.py
Normal file
180
docling/models/hf_vlm_model.py
Normal file
@@ -0,0 +1,180 @@
|
||||
import logging
|
||||
import time
|
||||
from pathlib import Path
|
||||
from typing import Iterable, List, Optional
|
||||
|
||||
from docling.datamodel.base_models import Page, VlmPrediction
|
||||
from docling.datamodel.document import ConversionResult
|
||||
from docling.datamodel.pipeline_options import (
|
||||
AcceleratorDevice,
|
||||
AcceleratorOptions,
|
||||
HuggingFaceVlmOptions,
|
||||
)
|
||||
from docling.datamodel.settings import settings
|
||||
from docling.models.base_model import BasePageModel
|
||||
from docling.utils.accelerator_utils import decide_device
|
||||
from docling.utils.profiling import TimeRecorder
|
||||
|
||||
_log = logging.getLogger(__name__)
|
||||
|
||||
|
||||
class HuggingFaceVlmModel(BasePageModel):
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
enabled: bool,
|
||||
artifacts_path: Optional[Path],
|
||||
accelerator_options: AcceleratorOptions,
|
||||
vlm_options: HuggingFaceVlmOptions,
|
||||
):
|
||||
self.enabled = enabled
|
||||
|
||||
self.vlm_options = vlm_options
|
||||
|
||||
if self.enabled:
|
||||
import torch
|
||||
from transformers import ( # type: ignore
|
||||
AutoModelForVision2Seq,
|
||||
AutoProcessor,
|
||||
BitsAndBytesConfig,
|
||||
)
|
||||
|
||||
device = decide_device(accelerator_options.device)
|
||||
self.device = device
|
||||
|
||||
_log.debug("Available device for HuggingFace VLM: {}".format(device))
|
||||
|
||||
repo_cache_folder = vlm_options.repo_id.replace("/", "--")
|
||||
|
||||
# PARAMETERS:
|
||||
if artifacts_path is None:
|
||||
artifacts_path = self.download_models(self.vlm_options.repo_id)
|
||||
elif (artifacts_path / repo_cache_folder).exists():
|
||||
artifacts_path = artifacts_path / repo_cache_folder
|
||||
|
||||
self.param_question = vlm_options.prompt # "Perform Layout Analysis."
|
||||
self.param_quantization_config = BitsAndBytesConfig(
|
||||
load_in_8bit=vlm_options.load_in_8bit, # True,
|
||||
llm_int8_threshold=vlm_options.llm_int8_threshold, # 6.0
|
||||
)
|
||||
self.param_quantized = vlm_options.quantized # False
|
||||
|
||||
self.processor = AutoProcessor.from_pretrained(artifacts_path)
|
||||
if not self.param_quantized:
|
||||
self.vlm_model = AutoModelForVision2Seq.from_pretrained(
|
||||
artifacts_path,
|
||||
device_map=device,
|
||||
torch_dtype=torch.bfloat16,
|
||||
_attn_implementation=(
|
||||
"flash_attention_2"
|
||||
if self.device.startswith("cuda")
|
||||
and accelerator_options.cuda_use_flash_attention2
|
||||
else "eager"
|
||||
),
|
||||
) # .to(self.device)
|
||||
|
||||
else:
|
||||
self.vlm_model = AutoModelForVision2Seq.from_pretrained(
|
||||
artifacts_path,
|
||||
device_map=device,
|
||||
torch_dtype="auto",
|
||||
quantization_config=self.param_quantization_config,
|
||||
_attn_implementation=(
|
||||
"flash_attention_2"
|
||||
if self.device.startswith("cuda")
|
||||
and accelerator_options.cuda_use_flash_attention2
|
||||
else "eager"
|
||||
),
|
||||
) # .to(self.device)
|
||||
|
||||
@staticmethod
|
||||
def download_models(
|
||||
repo_id: str,
|
||||
local_dir: Optional[Path] = None,
|
||||
force: bool = False,
|
||||
progress: bool = False,
|
||||
) -> Path:
|
||||
from huggingface_hub import snapshot_download
|
||||
from huggingface_hub.utils import disable_progress_bars
|
||||
|
||||
if not progress:
|
||||
disable_progress_bars()
|
||||
download_path = snapshot_download(
|
||||
repo_id=repo_id,
|
||||
force_download=force,
|
||||
local_dir=local_dir,
|
||||
# revision="v0.0.1",
|
||||
)
|
||||
|
||||
return Path(download_path)
|
||||
|
||||
def __call__(
|
||||
self, conv_res: ConversionResult, page_batch: Iterable[Page]
|
||||
) -> Iterable[Page]:
|
||||
for page in page_batch:
|
||||
assert page._backend is not None
|
||||
if not page._backend.is_valid():
|
||||
yield page
|
||||
else:
|
||||
with TimeRecorder(conv_res, "vlm"):
|
||||
assert page.size is not None
|
||||
|
||||
hi_res_image = page.get_image(scale=2.0) # 144dpi
|
||||
# hi_res_image = page.get_image(scale=1.0) # 72dpi
|
||||
|
||||
if hi_res_image is not None:
|
||||
im_width, im_height = hi_res_image.size
|
||||
|
||||
# populate page_tags with predicted doc tags
|
||||
page_tags = ""
|
||||
|
||||
if hi_res_image:
|
||||
if hi_res_image.mode != "RGB":
|
||||
hi_res_image = hi_res_image.convert("RGB")
|
||||
|
||||
messages = [
|
||||
{
|
||||
"role": "user",
|
||||
"content": [
|
||||
{
|
||||
"type": "text",
|
||||
"text": "This is a page from a document.",
|
||||
},
|
||||
{"type": "image"},
|
||||
{"type": "text", "text": self.param_question},
|
||||
],
|
||||
}
|
||||
]
|
||||
prompt = self.processor.apply_chat_template(
|
||||
messages, add_generation_prompt=False
|
||||
)
|
||||
inputs = self.processor(
|
||||
text=prompt, images=[hi_res_image], return_tensors="pt"
|
||||
)
|
||||
inputs = {k: v.to(self.device) for k, v in inputs.items()}
|
||||
|
||||
start_time = time.time()
|
||||
# Call model to generate:
|
||||
generated_ids = self.vlm_model.generate(
|
||||
**inputs, max_new_tokens=4096, use_cache=True
|
||||
)
|
||||
|
||||
generation_time = time.time() - start_time
|
||||
generated_texts = self.processor.batch_decode(
|
||||
generated_ids[:, inputs["input_ids"].shape[1] :],
|
||||
skip_special_tokens=False,
|
||||
)[0]
|
||||
|
||||
num_tokens = len(generated_ids[0])
|
||||
page_tags = generated_texts
|
||||
|
||||
# inference_time = time.time() - start_time
|
||||
# tokens_per_second = num_tokens / generation_time
|
||||
# print("")
|
||||
# print(f"Page Inference Time: {inference_time:.2f} seconds")
|
||||
# print(f"Total tokens on page: {num_tokens:.2f}")
|
||||
# print(f"Tokens/sec: {tokens_per_second:.2f}")
|
||||
# print("")
|
||||
page.predictions.vlm_response = VlmPrediction(text=page_tags)
|
||||
|
||||
yield page
|
||||
Reference in New Issue
Block a user