[Feature] Added num_envs parameter in GymEnv to call multiple environments in one call #3343
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3343
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New Failure, 1 Cancelled Job, 3 PendingAs of commit 2d34456 with merge base 736d6aa ( NEW FAILURE - The following job has failed:
CANCELLED JOB - The following job was cancelled. Please retry:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
meta-cla
bot
added
the
CLA Signed
ParamThakkar123
changed the title
vmoens
force-pushed
the
add/num-envs-gym
branch
from
305fe34 to
1a24ab5
Compare
vmoens
requested changes
Jan 19, 2026
torchrl/envs/libs/gym.py
Outdated
Comment on lines
1733
to
1734
| ``1`` (a single env is to be run). When ``num_envs > 1``, a :class:`~torchrl.envs.ParallelEnv` | ||
| will be returned. |
Collaborator
There was a problem hiding this comment.
This is bc-breaking, let's not do that. If a PEnv needs to be used let's let the users ask for it explicitly in some way
Contributor
Author
There was a problem hiding this comment.
Sure. Agree on this. I will keep it similar to my last PR
Collaborator
There was a problem hiding this comment.
Hmm not sure what you mean. Maybe we don't need it for gym since it already exists? Alternatively we can change the arg name to make it explicit it's torchrl-parallel (since it's nightly we can also change it in dm-control for consistency)
Contributor
Author
There was a problem hiding this comment.
Like what I actually meant was following the same structure or flow we did for DMControl but as you said that gym already has the feature for running multiple envs we can do it for other environments instead of this one
Contributor
Author
There was a problem hiding this comment.
And yes your latter idea for torchrl-parallel seems awesome and we can go ahead with it?
Collaborator
There was a problem hiding this comment.
let's just convene on a kwarg name?
num_envs: gym API
num_workers: torchrl?
Maybe not super explicit
Contributor
Author
There was a problem hiding this comment.
Yes. Seems good.
Contributor
Author
There was a problem hiding this comment.
Should I close this one in that case as gym envs already have num_envs parameter ? and start on the ones that don't have it ?
Collaborator
There was a problem hiding this comment.
I thought we said we'd keep the new feature but with a different name?
Contributor
Author
There was a problem hiding this comment.
Oh. sorry. I lost some track on that. Yes. I will keep the feature
vmoens
reviewed
Jan 19, 2026
torchrl/envs/libs/gym.py
Outdated
Comment on lines
1733
to
1734
| ``1`` (a single env is to be run). When ``num_envs > 1``, a :class:`~torchrl.envs.ParallelEnv` | ||
| will be returned. |
Collaborator
There was a problem hiding this comment.
I thought we said we'd keep the new feature but with a different name?
vmoens
reviewed
Jan 20, 2026
vmoens
reviewed
Jan 20, 2026
Contributor
Author
|
@vmoens All changes noted 🫡 |
Collaborator
from the CI |
Contributor
Author
|
@vmoens I fixed the failing tests |
Collaborator
|
From https://github.com/pytorch/rl/actions/runs/21169673684/job/60889590869?pr=3343 |
Contributor
Author
|
Tests give really weird errors on gym 0.19: |
Contributor
Author
|
Got similar error on |
Contributor
Author
|
The code I just pushed works for all versions except gym==0.19 and 0.20 maybe because of the windows compatibility error |
Contributor
Author
|
might work on CI but didn't work locally |
Collaborator
No it did not haha |
Contributor
Author
|
🥲🥲🥲 |
- Remove incorrect CUDA skip decorator (test doesn't use CUDA) - Use Pendulum-v1 instead of CartPole-v1 because CartPole can terminate early due to pole falling, especially with frame_skip=4, causing the rollout assertion to fail (expected 5 steps, got 3)
Use PENDULUM_VERSIONED() instead of hardcoded "Pendulum-v1" to support older gym versions that only have "Pendulum-v0".
3 tasks
- Fix import ordering - Add missing space after == - Add newline at end of file - Remove trailing whitespace
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
Describe your changes in detail.
Motivation and Context
Why is this change required? What problem does it solve?
If it fixes an open issue, please link to the issue here.
You can use the syntax
close #15213if this solves the issue #15213Types of changes
What types of changes does your code introduce? Remove all that do not apply:
Checklist
Go over all the following points, and put an
xin all the boxes that apply.If you are unsure about any of these, don't hesitate to ask. We are here to help!