NVIDIA / NVIDIA/NeMo-Retriever
[FEA]: Add batch support to extraction calls for paddle, cached, and deplot
Open
@drobison00 is already working on this.
Since Sep 6, 2024.
feature request
- Dominant language
- Python
- Stars
- 3k
- Forks
- 349
- Avg merge
- 1d 23h
- Merged PRs (30d)
- 116
Description
Is this a new feature, an improvement, or a change to existing functionality?
Change
How would you describe the priority of this feature request
Significant improvement
Please provide a clear description of problem this feature solves
The current pdfium extraction path uses batch size = 1 for calls to paddle, yolox, and cached; this needs to be updated so that we construct batches up to some threshold at the document level.
Describe the feature, and optionally a solution or implementation and any alternatives
This is straight forward and requires updates to handle_table_chart_extraction, and various NIM helper functions so that they are batch aware.
Additional context
No response
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Assessment
This issue has not been assessed yet.