Replies: 1 comment
-
|
Hi @isNotOkay! I'm Dosu and I’m helping the docling team. The Docling API (including the official Docker images) does not currently provide a confidence score for PDF processing results. The output is structured (paragraphs, tables, etc.), and confidence metrics from OCR or layout analysis are not exposed in the API response. This is a known feature request and has been discussed by maintainers, but it is not yet implemented—see the discussion and feature request for more details: docling/discussions/419, docling/issues/1102. If you need confidence estimation, you might consider using the Python library for more granular access or keep an eye on future updates to the API. Let me know if you want help with workarounds or extracting other metadata! To reply, just mention @dosu. How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other
|
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
Hey everyone,
I'd like to get the confidence score of a PDF document using the API. Is this possible? I need to know how confident a result is, because some PDFs contain low resolution scans, which usually leads to bad results.
I use the official docker images, not the python library.
Beta Was this translation helpful? Give feedback.
All reactions