You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I see that in robomimic wrapper. the done value is being automatically set as False. If we want to change it to its true value, is returning the actual done value enough ? Or are there some other changes in the training code that we need to handle for trajectories that terminate early ?
The text was updated successfully, but these errors were encountered:
Hi,
I see that in robomimic wrapper. the done value is being automatically set as False. If we want to change it to its true value, is returning the actual done value enough ? Or are there some other changes in the training code that we need to handle for trajectories that terminate early ?
The text was updated successfully, but these errors were encountered: