mirrors/docling
1
0
Fork 0
mirror of https://github.com/DS4SD/docling.git synced 2025-07-26 20:14:47 +00:00
Code Issues Packages Projects Releases Wiki Activity

refactoring the download_model

Browse Source 
Signed-off-by: Peter Staar <taa@zurich.ibm.com>
This commit is contained in:
Peter Staar 2025-05-14 05:31:54 +02:00
parent 3407955a47
commit 4c0bc61e54
9 changed files with 64 additions and 221 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
docling/datamodel/pipeline_options.py
View File

@ -269,7 +269,9 @@ class InferenceFramework(str, Enum):
OPENAI = "openai"
TRANSFORMERS_AutoModelForVision2Seq = "transformers-AutoModelForVision2Seq"
TRANSFORMERS_AutoModelForCausalLM = "transformers-AutoModelForCausalLM"
TRANSFORMERS_LlavaForConditionalGeneration = "transformers-LlavaForConditionalGeneration"
TRANSFORMERS_LlavaForConditionalGeneration = (
"transformers-LlavaForConditionalGeneration"
)
class HuggingFaceVlmOptions(BaseVlmOptions):

153
docling/models/hf_vlm_model.py
View File

@ -17,81 +17,7 @@ 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(f"Available device for HuggingFace VLM: {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,
# trust_remote_code=True,
)
if not self.param_quantized:
self.vlm_model = AutoModelForVision2Seq.from_pretrained(
artifacts_path,
device_map=self.device,
torch_dtype=torch.bfloat16,
_attn_implementation=(
"flash_attention_2"
if self.device.startswith("cuda")
and accelerator_options.cuda_use_flash_attention2
else "eager"
),
# trust_remote_code=True,
) # .to(self.device)
else:
self.vlm_model = AutoModelForVision2Seq.from_pretrained(
artifacts_path,
device_map=self.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"
),
# trust_remote_code=True,
) # .to(self.device)
"""
class HuggingFaceVlmModel:
@staticmethod
def download_models(
repo_id: str,
@ -112,80 +38,3 @@ class HuggingFaceVlmModel(BasePageModel):
)
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
_log.debug(
f"Generated {num_tokens} tokens in time {generation_time:.2f} seconds."
)
# 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
"""

8
docling/models/hf_vlm_models/hf_vlm_mlx_model.py
View File

@ -11,6 +11,7 @@ from docling.datamodel.pipeline_options import (
HuggingFaceVlmOptions,
)
from docling.models.base_model import BasePageModel
from docling.models.hf_vlm_model import HuggingFaceVlmModel
from docling.utils.profiling import TimeRecorder
_log = logging.getLogger(__name__)
@ -44,7 +45,10 @@ class HuggingFaceMlxModel(BasePageModel):
# PARAMETERS:
if artifacts_path is None:
artifacts_path = self.download_models(self.vlm_options.repo_id)
# artifacts_path = self.download_models(self.vlm_options.repo_id)
artifacts_path = HuggingFaceVlmModel.download_models(
self.vlm_options.repo_id
)
elif (artifacts_path / repo_cache_folder).exists():
artifacts_path = artifacts_path / repo_cache_folder
@ -54,6 +58,7 @@ class HuggingFaceMlxModel(BasePageModel):
self.vlm_model, self.processor = load(artifacts_path)
self.config = load_config(artifacts_path)
"""
@staticmethod
def download_models(
repo_id: str,
@ -74,6 +79,7 @@ class HuggingFaceMlxModel(BasePageModel):
)
return Path(download_path)
"""
def __call__(
self, conv_res: ConversionResult, page_batch: Iterable[Page]

9
docling/models/hf_vlm_models/hf_vlm_model_AutoModelForCausalLM.py
View File

@ -11,6 +11,7 @@ from docling.datamodel.pipeline_options import (
HuggingFaceVlmOptions,
)
from docling.models.base_model import BasePageModel
from docling.models.hf_vlm_model import HuggingFaceVlmModel
from docling.utils.accelerator_utils import decide_device
from docling.utils.profiling import TimeRecorder
@ -30,7 +31,6 @@ class HuggingFaceVlmModel_AutoModelForCausalLM(BasePageModel):
self.trust_remote_code = True
self.vlm_options = vlm_options
print(self.vlm_options)
if self.enabled:
import torch
@ -49,7 +49,10 @@ class HuggingFaceVlmModel_AutoModelForCausalLM(BasePageModel):
# PARAMETERS:
if artifacts_path is None:
artifacts_path = self.download_models(self.vlm_options.repo_id)
# artifacts_path = self.download_models(self.vlm_options.repo_id)
artifacts_path = HuggingFaceVlmModel.download_models(
self.vlm_options.repo_id
)
elif (artifacts_path / repo_cache_folder).exists():
artifacts_path = artifacts_path / repo_cache_folder
@ -99,6 +102,7 @@ class HuggingFaceVlmModel_AutoModelForCausalLM(BasePageModel):
# Load generation config
self.generation_config = GenerationConfig.from_pretrained(model_path)
"""
@staticmethod
def download_models(
repo_id: str,
@ -119,6 +123,7 @@ class HuggingFaceVlmModel_AutoModelForCausalLM(BasePageModel):
)
return Path(download_path)
"""
def __call__(
self, conv_res: ConversionResult, page_batch: Iterable[Page]

8
docling/models/hf_vlm_models/hf_vlm_model_AutoModelForVision2Seq.py
View File

@ -11,6 +11,7 @@ from docling.datamodel.pipeline_options import (
HuggingFaceVlmOptions,
)
from docling.models.base_model import BasePageModel
from docling.models.hf_vlm_model import HuggingFaceVlmModel
from docling.utils.accelerator_utils import decide_device
from docling.utils.profiling import TimeRecorder
@ -46,7 +47,10 @@ class HuggingFaceVlmModel_AutoModelForVision2Seq(BasePageModel):
# PARAMETERS:
if artifacts_path is None:
artifacts_path = self.download_models(self.vlm_options.repo_id)
# artifacts_path = self.download_models(self.vlm_options.repo_id)
artifacts_path = HuggingFaceVlmModel.download_models(
self.vlm_options.repo_id
)
elif (artifacts_path / repo_cache_folder).exists():
artifacts_path = artifacts_path / repo_cache_folder
@ -90,6 +94,7 @@ class HuggingFaceVlmModel_AutoModelForVision2Seq(BasePageModel):
# trust_remote_code=True,
) # .to(self.device)
"""
@staticmethod
def download_models(
repo_id: str,
@ -110,6 +115,7 @@ class HuggingFaceVlmModel_AutoModelForVision2Seq(BasePageModel):
)
return Path(download_path)
"""
def __call__(
self, conv_res: ConversionResult, page_batch: Iterable[Page]

40
docling/models/hf_vlm_models/hf_vlm_model_LlavaForConditionalGeneration.py
View File

@ -4,6 +4,8 @@ from collections.abc import Iterable
from pathlib import Path
from typing import Optional
from transformers import AutoProcessor, LlavaForConditionalGeneration
from docling.datamodel.base_models import Page, VlmPrediction
from docling.datamodel.document import ConversionResult
from docling.datamodel.pipeline_options import (
@ -11,11 +13,10 @@ from docling.datamodel.pipeline_options import (
HuggingFaceVlmOptions,
)
from docling.models.base_model import BasePageModel
from docling.models.hf_vlm_model import HuggingFaceVlmModel
from docling.utils.accelerator_utils import decide_device
from docling.utils.profiling import TimeRecorder
from transformers import AutoProcessor, LlavaForConditionalGeneration
_log = logging.getLogger(__name__)
@ -32,13 +33,12 @@ class HuggingFaceVlmModel_LlavaForConditionalGeneration(BasePageModel):
self.trust_remote_code = True
self.vlm_options = vlm_options
print(self.vlm_options)
if self.enabled:
import torch
from transformers import ( # type: ignore
LlavaForConditionalGeneration,
AutoProcessor,
LlavaForConditionalGeneration,
)
self.device = decide_device(accelerator_options.device)
@ -51,22 +51,27 @@ class HuggingFaceVlmModel_LlavaForConditionalGeneration(BasePageModel):
# PARAMETERS:
if artifacts_path is None:
artifacts_path = self.download_models(self.vlm_options.repo_id)
# artifacts_path = self.download_models(self.vlm_options.repo_id)
artifacts_path = HuggingFaceVlmModel.download_models(
self.vlm_options.repo_id
)
elif (artifacts_path / repo_cache_folder).exists():
artifacts_path = artifacts_path / repo_cache_folder
model_path = artifacts_path
print(f"model: {model_path}")
self.max_new_tokens = 64 # FIXME
self.max_new_tokens = 64 # FIXME
self.processor = AutoProcessor.from_pretrained(
artifacts_path,
trust_remote_code=self.trust_remote_code,
)
self.vlm_model = LlavaForConditionalGeneration.from_pretrained(artifacts_path).to(self.device)
self.vlm_model = LlavaForConditionalGeneration.from_pretrained(
artifacts_path
).to(self.device)
"""
@staticmethod
def download_models(
repo_id: str,
@ -87,6 +92,7 @@ class HuggingFaceVlmModel_LlavaForConditionalGeneration(BasePageModel):
)
return Path(download_path)
"""
def __call__(
self, conv_res: ConversionResult, page_batch: Iterable[Page]
@ -109,20 +115,22 @@ class HuggingFaceVlmModel_LlavaForConditionalGeneration(BasePageModel):
if hi_res_image.mode != "RGB":
hi_res_image = hi_res_image.convert("RGB")
images = [
hi_res_image
]
images = [hi_res_image]
prompt = "<s>[INST]Describe the images.\n[IMG][/INST]"
inputs = self.processor(text=prompt, images=images, return_tensors="pt", use_fast=False).to(self.device) #.to("cuda")
inputs = self.processor(
text=prompt, images=images, return_tensors="pt", use_fast=False
).to(self.device) # .to("cuda")
generate_ids = self.vlm_model.generate(
**inputs,
max_new_tokens=self.max_new_tokens,
use_cache=True # Enables KV caching which can improve performance
use_cache=True, # Enables KV caching which can improve performance
)
response = self.processor.batch_decode(generate_ids,
skip_special_tokens=True,
clean_up_tokenization_spaces=False)[0]
response = self.processor.batch_decode(
generate_ids,
skip_special_tokens=True,
clean_up_tokenization_spaces=False,
)[0]
print(f"response: {response}")
"""
_log.debug(

33
docling/models/hf_vlm_models/pixtral_12b_2409.py
View File

@ -1,33 +0,0 @@
import logging
import time
from collections.abc import Iterable
from pathlib import Path
from typing import Optional
from docling.datamodel.base_models import Page, VlmPrediction
from docling.datamodel.document import ConversionResult
from docling.datamodel.pipeline_options import (
AcceleratorOptions,
HuggingFaceVlmOptions,
)
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_pixtral_12b_2409(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

14
docling/pipeline/vlm_pipeline.py
View File

@ -24,18 +24,16 @@ from docling.datamodel.settings import settings
from docling.models.api_vlm_model import ApiVlmModel
# from docling.models.hf_vlm_model import HuggingFaceVlmModel
from docling.models.hf_vlm_models.hf_vlm_mlx_model import (
HuggingFaceMlxModel
)
from docling.models.hf_vlm_models.hf_vlm_model_LlavaForConditionalGeneration import (
HuggingFaceVlmModel_LlavaForConditionalGeneration
)
from docling.models.hf_vlm_models.hf_vlm_mlx_model import HuggingFaceMlxModel
from docling.models.hf_vlm_models.hf_vlm_model_AutoModelForCausalLM import (
HuggingFaceVlmModel_AutoModelForCausalLM,
)
from docling.models.hf_vlm_models.hf_vlm_model_AutoModelForVision2Seq import (
HuggingFaceVlmModel_AutoModelForVision2Seq,
)
from docling.models.hf_vlm_models.hf_vlm_model_LlavaForConditionalGeneration import (
HuggingFaceVlmModel_LlavaForConditionalGeneration,
)
from docling.pipeline.base_pipeline import PaginatedPipeline
from docling.utils.profiling import ProfilingScope, TimeRecorder
@ -126,7 +124,9 @@ class VlmPipeline(PaginatedPipeline):
),
]
else:
raise ValueError(f"Could not instantiate the right type of VLM pipeline: {vlm_options.inference_framework}")
raise ValueError(
f"Could not instantiate the right type of VLM pipeline: {vlm_options.inference_framework}"
)
self.enrichment_pipe = [
# Other models working on `NodeItem` elements in the DoclingDocument

8
docs/examples/minimal_vlm_pipeline.py
View File

@ -50,10 +50,10 @@ vlm_conversion_options = pixtral_vlm_conversion_options
"""
pixtral_vlm_conversion_options = HuggingFaceVlmOptions(
repo_id="mistral-community/pixtral-12b",
prompt="OCR this image and export it in MarkDown.",
response_format=ResponseFormat.MARKDOWN,
inference_framework=InferenceFramework.TRANSFORMERS_LlavaForConditionalGeneration,
repo_id="mistral-community/pixtral-12b",
prompt="OCR this image and export it in MarkDown.",
response_format=ResponseFormat.MARKDOWN,
inference_framework=InferenceFramework.TRANSFORMERS_LlavaForConditionalGeneration,
)
vlm_conversion_options = pixtral_vlm_conversion_options