Ignatii Dubyshkin
Ignatii Dubyshkin
Hello, I am, honestly, do not understand how the stride can be applied here. You can found the mathematical definition of shift operation [here](https://en.wikipedia.org/wiki/Shift_operator) for reference. So in case of...
>And I don't think sparse shift is efficient since the computation is also increasing. Sparse shift on forward pass(inference) is just memory copy(it's ZERO Flops) and should work very fast...
Hello. First, I am not the author any of mentioned articles and also I am not interesting to reproduce the original article results. I am using SSL(active_flag=False) for my own...
> > I will checked this part of code and try to fix it. > > Big thanks for you findings! > > Thank you for your reply! I will...
> > > > I will checked this part of code and try to fix it. > > > > Big thanks for you findings! > > > > >...
@xiexiaona I fixed bug with `active_flag=True`, so you can update and try again. Also, regarding reproducibility, I advice you to seek answers [here](https://github.com/DeadAt0m/ActiveSparseShifts-PyTorch/issues?q=is%3Aissue+is%3Aclosed).
> > @xiexiaona I fixed bug with `active_flag=True`, so you can update and try again. > > Also, regarding reproducibility, I advice you to seek answers [here](https://github.com/DeadAt0m/ActiveSparseShifts-PyTorch/issues?q=is%3Aissue+is%3Aclosed). > > Thank...
> > > > @xiexiaona I fixed bug with `active_flag=True`, so you can update and try again. > > > > Also, regarding reproducibility, I advice you to seek answers...
> > > > > > @xiexiaona I fixed bug with `active_flag=True`, so you can update and try again. > > > > > > Also, regarding reproducibility, I advice...
> @DeadAt0m Excuse me. Regarding reproducibility of Resnet-20, I think my model architecture is correct. Because my Resnet-20 achieves 91.97% accuracy in python SSL shift operation, which is matching to...