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
Right now, the conditioning example in train_conditioning.py is a red herring - the conditioning does nothing.
As suggested in a code review, we should make one of the conditions remove (at least one) of the modes in the hypergrid environment, so we can verify at test time that the conditioning is actually being used: #188 (comment)
This will require the conditioning to interact with the environment somehow which expands the scope of this a bit but should be interesting (and a high priority follow up to the original conditioning PR).
The text was updated successfully, but these errors were encountered:
Right now, the conditioning example in
train_conditioning.py
is a red herring - the conditioning does nothing.As suggested in a code review, we should make one of the conditions remove (at least one) of the modes in the hypergrid environment, so we can verify at test time that the conditioning is actually being used: #188 (comment)
This will require the conditioning to interact with the environment somehow which expands the scope of this a bit but should be interesting (and a high priority follow up to the original conditioning PR).
The text was updated successfully, but these errors were encountered: