MISSING_PAGE_FAIL: error on this pdf

gurusura commented 11 months ago

20.pdf I get the following error in Colab when I try it on the attached pdf /usr/local/lib/python3.10/dist-packages/torch/functional.py:504: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at ../aten/src/ATen/native/TensorShape.cpp:3483.) return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] 0% 0/1 [00:00<?, ?it/s]WARNING:root:Found repetitions in sample 0 INFO:root:Processing file 20.pdf with 1 pages WARNING:root:Skipping page 1 due to repetitions. 100% 1/1 [00:03<00:00, 3.71s/it]

Nougat seems to be unable to ocr simple division written in the format. Any way to overcome this? Thanks

KartavyaBagga commented 10 months ago

Did u find anything on this ?

gurusura commented 10 months ago

Still experimenting and comparing with other libraries. It doesn’t identify figures and photos and extract them; at least not for the pdf I tried. Must be the same for physics and chemistry diagrams

On Mon, 6 Nov 2023 at 10:07 PM, KartavyaBagga @.***> wrote:

Did u find anything on this ?

— Reply to this email directly, view it on GitHub https://github.com/facebookresearch/nougat/issues/152#issuecomment-1795411995, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAAV3CG6EWSPDWMTTZD7LU3YDEG5HAVCNFSM6AAAAAA6A2FEE6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTOOJVGQYTCOJZGU . You are receiving this because you authored the thread.Message ID: @.***>

facebookresearch / nougat

MISSING_PAGE_FAIL: error on this pdf #152