You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I wrapped docling in a fastapi service, and noticed that after running any of the endpoints that include convert that the cpu would spin up to convert doc, but something continues to run in the background (single core, 100% cpu -- no idea what it is).
...
Steps to reproduce
Run the convert function on a PDF doc then just pause execution. Run top command and you should see something lingering on the same process ID consuming a cpu core.
We usually support only the docling-serve REST API for Docling, we are trying to centralize all the work there. If you see that your use case is not covered or planned, we are happy to know about it.
I suggest you reopen the issue if you notice the same behavior in docling-serve.
Bug
I wrapped docling in a fastapi service, and noticed that after running any of the endpoints that include convert that the cpu would spin up to convert doc, but something continues to run in the background (single core, 100% cpu -- no idea what it is).
...
Steps to reproduce
Run the convert function on a PDF doc then just pause execution. Run top command and you should see something lingering on the same process ID consuming a cpu core.
Can be seen running this: https://github.com/patw/docinator
...
Docling version
docling 2.18.0
docling-core 2.17.2
docling-ibm-models 3.3.1
docling-parse 3.1.0
...
Python version
3.10.12 running under Ubuntu 22.04
...
The text was updated successfully, but these errors were encountered: