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 was trying to run the IBRL baseline for Can and Square tasks of robomimic and I got the following error -
Error in call to target 'agent.dataset.sequence.StitchedSequenceQLearningDataset':
KeyError('rewards is not a file in the archive')
full_key: offline_dataset
It looks like the train.npz file for can and square only has these fields ['states', 'actions', 'traj_lengths']
But Can-Ph and Square-Ph have all the fields - ['states', 'actions', 'rewards', 'terminals', 'traj_lengths']
The text was updated successfully, but these errors were encountered:
Hi,
I was trying to run the IBRL baseline for Can and Square tasks of robomimic and I got the following error -
It looks like the
train.npz
file for can and square only has these fields['states', 'actions', 'traj_lengths']
But Can-Ph and Square-Ph have all the fields -
['states', 'actions', 'rewards', 'terminals', 'traj_lengths']
The text was updated successfully, but these errors were encountered: