No DocTagsDocument from the example
#29
by
Lethnis
- opened
I tried image inference from the example, but there is no DocTagsDocument, so i stuck after this step
doctags = processor.batch_decode(
trimmed_generated_ids,
skip_special_tokens=False,
)[0].lstrip()
# can't do this
doctags_doc = DocTagsDocument.from_doctags_and_image_pairs([doctags], [image])
print(doctags)
# create a docling document
doc = DoclingDocument(name="Document")
doc.load_from_doctags(doctags_doc)
I've got the output, that looks like this <doctag><picture><loc_135><loc_20><loc_425><loc_348><other></picture>\n<text>.... So what can i do with it?
versions:
docling_core ==3.0.0
Realized it, after the second i made this post. I should've installed docling, not docling_core, though this is what example says
Installing docling_core is correct, however there is no version 3.0.0