kotaemon icon indicating copy to clipboard operation
kotaemon copied to clipboard

[REQUEST] - docling

Open thistleknot opened this issue 1 year ago • 1 comments

Reference Issues

No response

Summary

docling supports automatic parsing of pdf's with tables. I've found it very beneficial. https://github.com/DS4SD/docling/issues

Basic Example

automatic table extraction

Drawbacks

gpu access changes format of incoming document, but I've found it much easier to read pdfs processed by markdown. Uses layout detection + vision transformers to translate tables to markdown representations

Additional information

No response

thistleknot avatar Sep 16 '24 02:09 thistleknot

Thanks for the request. I'm working on it and will add it to the readers soon

cin-albert avatar Sep 26 '24 04:09 cin-albert