[RLlib; Docs overhaul] Docstring cleanup: Environments. #19784
Conversation
…_docstring_cleanup_envs
rllib/env/base_env.py
Outdated
|
|
||
| Args: | ||
| env: The environment type to convert/wrap. | ||
| make_env: A callable taking an int as input (number of sub-envs) |
There was a problem hiding this comment.
this reads a bit weird, return a list of sub-envs of type env?
There was a problem hiding this comment.
Haha, yeah, I'll try to clarify a bit better.
rllib/env/base_env.py
Outdated
| env = _ExternalEnvToBaseEnv(env) | ||
| elif isinstance(env, VectorEnv): | ||
| env = _VectorEnvToBaseEnv(env) | ||
| # `env` is already a BaseEnv -> Return as is. |
There was a problem hiding this comment.
this PR does have some logic changes?
There was a problem hiding this comment.
Nope, I just cleaned up the if block with a "early-out" -> Less indentation.
rllib/env/env_context.py
Outdated
| vector_index: When there are multiple envs per worker, this | ||
| uniquely identifies the env index within the worker. | ||
| Starts from 0. | ||
| remote: Whether single sub-environments (in a vectorized env) |
There was a problem hiding this comment.
Maybe individual sub-environments (...) sounds a bit better?
There was a problem hiding this comment.
Yeah, I should probably explain better, what a sub-env is :)
Aka "underlying" envs -> The n single (cloned) envs within a vectorized one.
I'll fix this. We should use the same terminology throughout (I like sub-envs as it hints that there are many sub-envs within the actual env, rather than "unwrapped" which could be interpreted as a single env that's just wrapped and there is no vectorization).
There was a problem hiding this comment.
I'm using "sub environments" now across the board, sometimes: "underlying sub environments". Also I added explanations as to what that means and made sure it's changes everywhere we used to have "underlying" or "wrapped". I also changed the method: get_underlying into get_sub_environment using soft-deprecation.
| """Initializes a RemoteVectorEnv instance. | ||
|
|
||
| Args: | ||
| make_env: Callable that produces a single (non-vectorized) env, |
There was a problem hiding this comment.
this is interesting. so here we only want make_env to return a single non-vectoriezed env? not like above.
…_docstring_cleanup_envs
|
Hey @gjoliver , could you take another look? I addressed all your questions and change requests. |
…_docstring_cleanup_envs
…_docstring_cleanup_envs # Conflicts: # rllib/agents/callbacks.py
…otherwise, but seem to fail nevertheless with this PR's changes.
…_docstring_cleanup_envs
…_docstring_cleanup_envs
Docstring cleanup: rllib/env folder.
Why are these changes needed?
Related issue number
Checks
scripts/format.shto lint the changes in this PR.