Closed Aakarshan-chauhan closed 3 years ago
For environments with action shape ( 1 , ) , the action values are in a 1D array instead of being in batches. Minor changes for fixing it
( 1 , )
Hey, thanks for the pull request. But, to make it more explicit and not affect performance on other environments, change it to :
if(self.action_dim == 1): action = action.reshape(-1, 1)
For environments with action shape
( 1 , )
, the action values are in a 1D array instead of being in batches. Minor changes for fixing it