Artem Zholus
Artem Zholus
same issue
Hi @LYK-love . We did all the logging for our work in wandb so I can export them and upload to our repo within the same format as DreamerV3 did....
Looks like this is a bug. I will take a look.
Hi @Liuxueyi . Apologies for not getting back to you on this sooner. Both first authors are currently attending ICLR 24 where we present this paper. I will get back...
Hi @Liuxueyi . the sentence you quoted is a general recommendation for working with new environments. There is another part of the paper where we state that in popgym we...
This looks like a serious issue. I will take a look.
Hi @LYK-love , after a quick search I found that the most likely reason is the mismatch of SSM hyperparameters. For example, you are probably using the hidden size of...
Please note that the hidden size alone does not guarantee to fix everything, but you can try that in the meantime.
Hi @YJ-142150 ! We don't plan to release the training code but we instead released the pytorch code to compute the tapnext loss and compute it's gradient. the code is...
Hi @bhack, I plan to work on this, but not as part of Google. This will likely take a couple of weeks.