docling-project / docling-project/docling
Pipeline options to generate images from Non-PDF files
- Dominant language
- Python
- Stars
- 66.4k
- Forks
- 4.8k
- Avg merge
- 2d 21h
- Merged PRs (30d)
- 84
Description
Hi - I am trying to use docling to convert files that are not pdfs or images i.e. word, ppt, websites, excel, etc. Some of these files have images embedded in them. When I use the docling pipeline to covert PDFs, I can use PDF pipeline options. When I choose not to specify the file input type and allow docling to choose to best document conversion strategy:
`
converter = DocumentConverter()
doc = converter.convert(source).document
`
I am not able to specify pipeline options such as generate picture images or accelerator options.
Is it possible to create an omnibus pipeline options that applies to all possible input types when allowing docling to automatically choose how to convert? Thanks!
Contributor guide
Assessment
This issue has not been assessed yet.