mirrors/docling
1
0
Fork 0
mirror of https://github.com/DS4SD/docling.git synced 2025-12-08 20:58:11 +00:00
Code Issues Packages Projects Releases Wiki Activity

feat: Support cuda:n GPU device allocation (#694)

Browse Source 
* Adding multi-gpu support, and cuda device allocation

Signed-off-by: ahn <ahn@zurich.ibm.com>

* Fixes pydantic exception with cuda:n
Signed-off-by: ahn <ahn@zurich.ibm.com>

* Pydantic field validator and comment restored.

Signed-off-by: ahn <ahn@zurich.ibm.com>

* chore: Accept AcceleratorDevice enum type

Signed-off-by: Christoph Auer <cau@zurich.ibm.com>

* Resetted some options to default, removed EasyOCR model wrap.
Signed-off-by: ahn <ahn@zurich.ibm.com>

* Fixed rebased issues
Signed-off-by: ahn <ahn@zurich.ibm.com>

* Revert accelerator test options
Signed-off-by: ahn <ahn@zurich.ibm.com>

---------

Signed-off-by: ahn <ahn@zurich.ibm.com>
Signed-off-by: Christoph Auer <cau@zurich.ibm.com>
Co-authored-by: ahn <ahn@sonny.zuvela.ibm.com>
Co-authored-by: ahn <ahn@zurich.ibm.com>
Co-authored-by: Christoph Auer <cau@zurich.ibm.com>
This commit is contained in:
Ahmed Nassar
2025-02-17 11:31:13 +01:00
committed by GitHub
parent 428b656793
commit 77eb77bdc2
3 changed files with 73 additions and 19 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
docs/examples/run_with_accelerator.py
View File

@@ -30,6 +30,9 @@ def main():
# num_threads=8, device=AcceleratorDevice.CUDA
# )
# easyocr doesnt support cuda:N allocation, defaults to cuda:0
# accelerator_options = AcceleratorOptions(num_threads=8, device="cuda:1")
pipeline_options = PdfPipelineOptions()
pipeline_options.accelerator_options = accelerator_options
pipeline_options.do_ocr = True