search

ibm-aur-nlp / PubTabNet

Other
380 stars 79 forks source link

Dataset Size #20

Open ctensmeyer opened 3 years ago

ctensmeyer commented 3 years ago

In the ECCV 2020 paper, the size of PubTabNet is said to be 568K images + HTML. When I downloaded the v2 PubTabNet, I count 519030 image files and 509892 lines in the PubTabNet_2.0.0.jsonl file (presumably, the test split GT is omitted).

Can you explain this discrepancy and what effect it has on the ability of researchers to reproduce or fairly compare with your ECCV paper?

ajjimeno commented 3 years ago

Hi Christopher, As you mention there is a subset that is being used in the following challenge as test set: https://icdar2021.org/competitions/competition-on-scientific-literature-parsing/

On Wed, Feb 3, 2021 at 3:28 AM Christopher Tensmeyer < notifications@github.com> wrote:

In the ECCV 2020 paper, the size of PubTabNet is said to be 568K images + HTML. When I downloaded the v2 PubTabNet, I count 519030 image files and 509892 lines in the PubTabNet_2.0.0.jsonl file (presumably, the test split GT is omitted).

Can you explain this discrepancy and what effect it has on the ability of researchers to reproduce or fairly compare with your ECCV paper?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/ibm-aur-nlp/PubTabNet/issues/20, or unsubscribe https://github.com/notifications/unsubscribe-auth/AA6BZDPFJT4BRX6F5YEDZ7TS5AR4VANCNFSM4W67GCNQ .

ctensmeyer commented 3 years ago

After the competition is completed, will the full 568K dataset be released to allow comparison with the ECCV paper?

ajjimeno commented 3 years ago

That is the intention. There is time till the 31st of March to participate in the challenge, in case you are interested.

On Fri, Feb 5, 2021 at 11:28 AM Christopher Tensmeyer < notifications@github.com> wrote:

After the competition is completed, will the full 568K dataset be released to allow comparison with the ECCV paper?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/ibm-aur-nlp/PubTabNet/issues/20#issuecomment-773694354, or unsubscribe https://github.com/notifications/unsubscribe-auth/AA6BZDIA4RTI6XMGTVPJ25TS5M3RJANCNFSM4W67GCNQ .