Skip to content

Commit

Permalink
fix: allow mps usage for easyocr (#286)
Browse files Browse the repository at this point in the history
* fix: allow mps usage for easyocr

Signed-off-by: Michele Dolfi <[email protected]>

* add example for cpu-only

Signed-off-by: Michele Dolfi <[email protected]>

* comment out example

Signed-off-by: Michele Dolfi <[email protected]>

---------

Signed-off-by: Michele Dolfi <[email protected]>
  • Loading branch information
dolfim-ibm authored Nov 10, 2024
1 parent be8aa17 commit 97f214e
Show file tree
Hide file tree
Showing 2 changed files with 15 additions and 4 deletions.
5 changes: 1 addition & 4 deletions docling/models/easyocr_model.py
Original file line number Diff line number Diff line change
Expand Up @@ -31,12 +31,9 @@ def __init__(self, enabled: bool, options: EasyOcrOptions):
"Alternatively, Docling has support for other OCR engines. See the documentation."
)

use_gpu = (
False if torch.backends.mps.is_available() else self.options.use_gpu
)
self.reader = easyocr.Reader(
lang_list=self.options.lang,
gpu=use_gpu,
gpu=self.options.use_gpu,
model_storage_directory=self.options.model_storage_directory,
download_enabled=self.options.download_enabled,
)
Expand Down
14 changes: 14 additions & 0 deletions docs/examples/custom_convert.py
Original file line number Diff line number Diff line change
Expand Up @@ -80,6 +80,20 @@ def main():
}
)

# Docling Parse with EasyOCR (CPU only)
# ----------------------
# pipeline_options = PdfPipelineOptions()
# pipeline_options.do_ocr = True
# pipeline_options.ocr_options.use_gpu = False # <-- set this.
# pipeline_options.do_table_structure = True
# pipeline_options.table_structure_options.do_cell_matching = True

# doc_converter = DocumentConverter(
# format_options={
# InputFormat.PDF: PdfFormatOption(pipeline_options=pipeline_options)
# }
# )

# Docling Parse with Tesseract
# ----------------------
# pipeline_options = PdfPipelineOptions()
Expand Down

0 comments on commit 97f214e

Please sign in to comment.