update locomotion track length #235
Conversation
There was a problem hiding this comment.
Could you elaborate more on the normalize_act, part? I will include the paragraph here:
The two environments are myoChallengeOslRunRandom-v0 and myoChallengeOslRunFixed-v0 can be accessed via env = gym.make(“myoChallengeOslRunRandom-v0”).
Both environments can be customized for ML or non-ML usage. For participants using ML-based methods, the action space can be set to between [-1 to 1] for both training and your submission with the normalize_act argument during environment creation. For participants using non-ML based methods, setting normalize_act=False would provide you with the muscle action space to be between [0 to 1]
During training, you can set this option with env = gym.make(env_name, normalize_act=True) for the action space [-1 to 1] and normalize_act=False for action space [0 to 1]
More information about environment customization during submission can be found in the submission file
update locomotion track length
No description provided.