mirrors/docling
1
0
Fork 0
mirror of https://github.com/DS4SD/docling.git synced 2025-07-27 20:44:16 +00:00
Code Issues Packages Projects Releases Wiki Activity

Moved artifacts_path for SmolDocling into vlm_options instead of global pipeline option

Browse Source 
Signed-off-by: Maksym Lysak <mly@zurich.ibm.com>
This commit is contained in:
Maksym Lysak 2025-01-21 18:00:05 +01:00
parent f2751e11f9
commit 7c4ab5c716
3 changed files with 4 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
docling/models/smol_docling_model.py
View File

@ -30,7 +30,7 @@ class SmolDoclingModel(BasePageModel):
def __init__( def __init__(
self, self,
artifacts_path: Path, # artifacts_path: Path,
accelerator_options: AcceleratorOptions, accelerator_options: AcceleratorOptions,
vlm_options: SmolDoclingOptions, vlm_options: SmolDoclingOptions,
): ):
@ -39,6 +39,7 @@ class SmolDoclingModel(BasePageModel):
_log.info("Available device for SmolDocling: {}".format(device)) _log.info("Available device for SmolDocling: {}".format(device))
# PARAMETERS: # PARAMETERS:
artifacts_path = Path(vlm_options.artifacts_path)
self.param_question = vlm_options.question # "Perform Layout Analysis." self.param_question = vlm_options.question # "Perform Layout Analysis."
self.param_quantization_config = BitsAndBytesConfig( self.param_quantization_config = BitsAndBytesConfig(
load_in_8bit=vlm_options.load_in_8bit, # True, load_in_8bit=vlm_options.load_in_8bit, # True,

3
docling/pipeline/vlm_pipeline.py
View File

@ -45,10 +45,12 @@ class VlmPipeline(PaginatedPipeline):
# force_backend_text = True - get text from backend using bounding boxes predicted by SmolDoclingss # force_backend_text = True - get text from backend using bounding boxes predicted by SmolDoclingss
self.force_backend_text = pipeline_options.force_backend_text self.force_backend_text = pipeline_options.force_backend_text
"""
if pipeline_options.artifacts_path is None: if pipeline_options.artifacts_path is None:
self.artifacts_path = self.download_models_hf() self.artifacts_path = self.download_models_hf()
else: else:
self.artifacts_path = Path(pipeline_options.artifacts_path) self.artifacts_path = Path(pipeline_options.artifacts_path)
"""
keep_images = ( keep_images = (
self.pipeline_options.generate_page_images self.pipeline_options.generate_page_images
@ -58,7 +60,6 @@ class VlmPipeline(PaginatedPipeline):
self.build_pipe = [ self.build_pipe = [
SmolDoclingModel( SmolDoclingModel(
artifacts_path=self.artifacts_path,
accelerator_options=pipeline_options.accelerator_options, accelerator_options=pipeline_options.accelerator_options,
vlm_options=self.pipeline_options.vlm_options, vlm_options=self.pipeline_options.vlm_options,
), ),

1
docs/examples/minimal_smol_docling.py
View File

@ -22,7 +22,6 @@ pipeline_options = PdfPipelineOptions()
pipeline_options.generate_page_images = True pipeline_options.generate_page_images = True
# If force_backend_text = True, text from backend will be used instead of generated text # If force_backend_text = True, text from backend will be used instead of generated text
pipeline_options.force_backend_text = False pipeline_options.force_backend_text = False
pipeline_options.artifacts_path = "model_artifacts/SmolDocling_2.7_DT_0.7"
vlm_options = SmolDoclingOptions( vlm_options = SmolDoclingOptions(
artifacts_path="model_artifacts/SmolDocling_2.7_DT_0.7", artifacts_path="model_artifacts/SmolDocling_2.7_DT_0.7",