Replies: 1 comment 1 reply
-
|
@laxas This would be definitely a useful addition, there is a work-in-progress for this feature here: DS4SD/docling-core#63 (contributed by @yasyf) |
Beta Was this translation helpful? Give feedback.
1 reply
-
|
That would definitely be an improvement. Including the page number in the chunk would be even better, as chunks could span across page breaks. Splitting pages before chunking could, in some cases, lead to two chunks where there should be one. |
Beta Was this translation helpful? Give feedback.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Is it possible (at least for PDFs) to preserve the information on which page (page number) a chunk startet?
I was considering chunkig single pages but found no methode to extract markdown.
Beta Was this translation helpful? Give feedback.
All reactions