Julieta Martinez
Julieta Martinez
Hi @meijieru, No I haven't solved this problem yet -- I haven't had time to debug that one number in our paper. I also cannot run this things right now...
@Nicholasli1995 Oh maybe this could explain the discrepancy. Our code checks for names correspondences with the H3.6M skeleton as it needs those to permute the data -- have you tried...
@Nicholasli1995 Upon re-reading your comment, I think you are referring to another table and another experiment. Could you please open another issue for that? Cheers,
@Nicholasli1995 I do not maintain the Pytorch implementation. Please open an issue in the corresponding repository. The expected number should be the one in the paper.
Nice catch @ohjay! I'll correct this when I get a chance.
Hi @hacchi-kh, This is a great find! It might explain why I was never able to reproduce certain numbers in the paper (issue https://github.com/una-dinosauria/3d-pose-baseline/issues/60) -- cc @matteorr. Thanks for reporting...
Huh. Can you please share the entire log? Is this with a single GPU?
Yep that looks like the log, but definitely not all of it.
It seems like the accuracy after the initial compression is too low (5%). I'll try to reproduce on my end, thanks for bringing this up!
Are you seeing this with other models too? Or just with Resnet 18? Could any of you provide a docker image to reproduce your error? (I should have provided one...