How to get observation space, or at least low/high bounds?

[ikamensh]

Hi,

I'm used to having observation_space in gym.Env / VectorEnv classes that tell me how to normalize observations, among other uses. Am I right to think this is missing in brax?

I'm doing

from brax import envs as brax_envs
vecenv = brax_envs.create("halfcheetah", batch_size=training_num)

and I can't find any property along the lines of observation space in the vecenv object

[erikfrey]

Hello there,

You can get these properties if you use the gym wrapper:

from brax.envs.wrappers import gym
env = gym.VectorGymWrapper(vecenv)
print(env.observation_space)

[btaba]

Hi @ikamensh , while the observation_space is populated, they're set to inf by default

brax/brax/envs/wrappers/gym.py

Line 114 in 3b72c95

obs = np.inf * np.ones(self._env.observation_size, dtype='float32')

If you'd like to make a PR to populate it generically, you can use dof limits

brax/brax/base.py

Line 294 in 3b72c95

limit: Tuple[jp.ndarray, jp.ndarray]