Skip to content

feat: add use_bundled_models flag to RapidOcrOptions (#2584) #2896

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
geoHeil wants to merge 1 commit into
base: main
Choose a base branch
from
+29 −10
Draft

feat: add use_bundled_models flag to RapidOcrOptions (#2584) #2896

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 6 additions & 0 deletions docling/datamodel/pipeline_options.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -153,6 +153,12 @@ class RapidOcrOptions(OcrOptions):
rec_font_path: Optional[str] = None # Deprecated, please use font_path instead
font_path: Optional[str] = None # same default as rapidocr

# Use RapidOCR's bundled models (shipped with the package) instead of artifacts_path
# When True: Ignores artifacts_path and uses models from site-packages/rapidocr/models
# When False (default): Follows Docling's standard behavior - uses artifacts_path when set
# Note: Explicitly set model paths (det_model_path, etc.) always take precedence
use_bundled_models: bool = False

# Dictionary to overwrite or pass-through additional parameters
rapidocr_params: Dict[str, Any] = Field(default_factory=dict)

Expand Down
33 changes: 23 additions & 10 deletions docling/models/stages/ocr/rapid_ocr_model.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -134,7 +134,11 @@ def __init__(
rec_model_path = self.options.rec_model_path
rec_keys_path = self.options.rec_keys_path
font_path = self.options.font_path
if artifacts_path is not None:

# Handle model path resolution based on use_bundled_models flag
# When use_bundled_models=True: Skip artifacts_path and let RapidOCR use bundled models
# When use_bundled_models=False (default): Follow Docling's standard behavior
if not self.options.use_bundled_models and artifacts_path is not None:
det_model_path = (
det_model_path
or artifacts_path
Expand Down Expand Up @@ -178,37 +182,46 @@ def __init__(
if not Path(model_path).exists():
_log.warning(f"The provided model path {model_path} is not found.")

# Build params dict with base settings
params = {
# Global settings (these are still correct)
# Global settings
"Global.text_score": self.options.text_score,
"Global.font_path": font_path,
# "Global.verbose": self.options.print_verbose,
# Detection model settings
"Det.model_path": det_model_path,
"Det.use_cuda": use_cuda,
"Det.use_dml": use_dml,
"Det.intra_op_num_threads": intra_op_num_threads,
"Det.engine_type": backend_enum,
# Classification model settings
"Cls.model_path": cls_model_path,
"Cls.use_cuda": use_cuda,
"Cls.use_dml": use_dml,
"Cls.intra_op_num_threads": intra_op_num_threads,
"Cls.engine_type": backend_enum,
# Recognition model settings
"Rec.model_path": rec_model_path,
"Rec.font_path": font_path,
"Rec.rec_keys_path": rec_keys_path,
"Rec.use_cuda": use_cuda,
"Rec.use_dml": use_dml,
"Rec.intra_op_num_threads": intra_op_num_threads,
"Det.engine_type": backend_enum,
"Cls.engine_type": backend_enum,
"Rec.engine_type": backend_enum,
"EngineConfig.paddle.use_cuda": use_cuda,
"EngineConfig.paddle.gpu_id": gpu_id,
"EngineConfig.torch.use_cuda": use_cuda,
"EngineConfig.torch.gpu_id": gpu_id,
}

# Only add model paths if they are explicitly set (not None)
# This allows RapidOCR to use its bundled models when use_bundled_models=True
if det_model_path is not None:
params["Det.model_path"] = det_model_path
if cls_model_path is not None:
params["Cls.model_path"] = cls_model_path
if rec_model_path is not None:
params["Rec.model_path"] = rec_model_path
if rec_keys_path is not None:
params["Rec.rec_keys_path"] = rec_keys_path
if font_path is not None:
params["Global.font_path"] = font_path
params["Rec.font_path"] = font_path

if self.options.rec_font_path is not None:
_log.warning(
"The 'rec_font_path' option for RapidOCR is deprecated. Please use 'font_path' instead."
Expand Down