Open elle-miller opened 5 days ago
Hi @elle-miller
Yes, in latest release (version 1.2.0) the automatic detection of the environment to be wrapped was updated and it didn't include DeepMind envs.
Regarding the AttributeError
error when wrapping the DeepMind environment, the DeepMind wrapper is not maintained anymore and will be removed in next version in favor of Shimmy. See skrl's Shimmy examples for training using a DeepMind environment.
Description
Hi there,
I have been using
skrl
with OIGE, but when I try the "Getting Started" code fordm_control
:I get the following error:
This is with Python 3.10.14,
dm_control==1.0.20
andskrl==1.1.0
. It seems that in the inheritance ofDeepMindWrapper
from in the baseWrapper
class,self._env.action_space
is being called beforeself._spec_to_space
can happen.Also just as a heads up, when I try with
skrl==1.2.0
, I now get this:Thank you!
What skrl version are you using?
1.1.0
What ML framework/library version are you using?
torch==2.3.1
Additional system information
Python 3.10.14