thuml/predrnn-pytorch

Whether the Moving_MNIST dataset volume in PredRNN++ is the same as the code repo?

ILoveStudying opened this issue · 3 comments

ILoveStudying commented

'Hi, this code repo is correct. We will rephrase the paper soon.

Originally posted by @wuhaixu2016 in #26 (comment)'

Thanks for your reply. I couldn't reproduce the performance on Moving_MNIST in PredRNN++ anyway. So I would like to know whether the Moving_MNIST dataset volume in PredRNN++ is the same as the code repo? If not, which volume setting is the experiment in PredRNN++ based on?
Looking forward to your reply. Thanks very much!

wuhaixu2016 commented

Hi, you can get the dataset and code here ⬇️
https://github.com/Yunbo426/predrnn-pp

ILoveStudying commented

Yes, I find that the Moving_MNIST dataset provided in code repo PredRNN++ is still inconsistent with the paper (PredRNN++). So which volume setting is the experiment in PredRNN++ based on?
Thanks!

wuhaixu2016 commented

I think you can totally follow the dataset since PredRNN, PredRNN++, PredRNN-V2 use the same Moving_mnist data.