Fixes setting the device from CLI in the RL training scripts #1013
Conversation
This file contains 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
amrmousa144
requested review from
jsmith-bdai,
Dhoeller19,
kellyguo11 and
Mayankm96
as code owners
amrmousa144
force-pushed
the
fix/wrong_device
branch
from
6e36e52 to
c57d5a1
Compare
Mayankm96
changed the title
6 tasks
Dhoeller19
reviewed
Sep 23, 2024
Dhoeller19
reviewed
Sep 23, 2024
Dhoeller19
reviewed
Sep 23, 2024
| @@ -120,7 +120,8 @@ def main(env_cfg: ManagerBasedRLEnvCfg | DirectRLEnvCfg | DirectMARLEnvCfg, agen | |||
| # set the environment seed | |||
| # note: certain randomizations occur in the environment initialization so we set the seed here | |||
| env_cfg.seed = args_cli.seed if args_cli.seed is not None else agent_cfg["seed"] | |||
|
|
|||
| env_cfg.sim.device = args_cli.device | |||
There was a problem hiding this comment.
Can you move it up to before if args_cli.distributed
and replace with env_cfg.sim.device = args_cli.device if args_cli.device is not None else env_cfg.sim.device
There was a problem hiding this comment.
Also need to remove L139-142 (multi-gpu training config). It is a duplication of L-110-111.
|
Thanks a lot for the fix! |
Co-authored-by: David Hoeller <[email protected]> Signed-off-by: Mayank Mittal <[email protected]>
|
Since it is a critical fix, I am merging this right now. Will push the remaining changes directly. |
Mayankm96
pushed a commit
that referenced
this pull request
Sep 24, 2024
This pull request fixes the issue where the device (`CPU` or `CUDA`) is not set correctly when using the `--device` argument in Hydra-configured scripts like `rsl_rl/train.py` and `skrl/train.py`. The bug caused the scripts to always default to `cuda:0`, even when `cpu` or a specific CUDA device (e.g., `cuda:1`) was selected. The fix adds the following line to ensure that the selected device is properly set in `env_cfg` before initializing the environment with `gym.make()`: ```python env_cfg.sim.device = args_cli.device ``` Fixes #1012 - Bug fix (non-breaking change which fixes an issue) Before: - skrl/train, when running the script with --device cpu, it defaults to cuda:0. - rsl_rl/train.py, the script freezes at `[INFO]: Starting the simulation. This may take a few seconds. Please wait....` After: - Both scripts run correctly on the specified device (e.g., cpu or cuda:1) without defaulting to cuda:0 or freezing. - [x] I have run the [`pre-commit` checks](https://pre-commit.com/) with `./isaaclab.sh --format` - [ ] I have made corresponding changes to the documentation - [x] My changes generate no new warnings - [ ] I have added tests that prove my fix is effective or that my feature works - [ ] I have updated the changelog and the corresponding version in the extension's `config/extension.toml` file - [x] I have added my name to the `CONTRIBUTORS.md` or my name already exists there
iamdrfly
pushed a commit
to iamdrfly/IsaacLab
that referenced
this pull request
Nov 21, 2024
…im#1013) This pull request fixes the issue where the device (`CPU` or `CUDA`) is not set correctly when using the `--device` argument in Hydra-configured scripts like `rsl_rl/train.py` and `skrl/train.py`. The bug caused the scripts to always default to `cuda:0`, even when `cpu` or a specific CUDA device (e.g., `cuda:1`) was selected. The fix adds the following line to ensure that the selected device is properly set in `env_cfg` before initializing the environment with `gym.make()`: ```python env_cfg.sim.device = args_cli.device ``` Fixes isaac-sim#1012 - Bug fix (non-breaking change which fixes an issue) Before: - skrl/train, when running the script with --device cpu, it defaults to cuda:0. - rsl_rl/train.py, the script freezes at `[INFO]: Starting the simulation. This may take a few seconds. Please wait....` After: - Both scripts run correctly on the specified device (e.g., cpu or cuda:1) without defaulting to cuda:0 or freezing. - [x] I have run the [`pre-commit` checks](https://pre-commit.com/) with `./isaaclab.sh --format` - [ ] I have made corresponding changes to the documentation - [x] My changes generate no new warnings - [ ] I have added tests that prove my fix is effective or that my feature works - [ ] I have updated the changelog and the corresponding version in the extension's `config/extension.toml` file - [x] I have added my name to the `CONTRIBUTORS.md` or my name already exists there
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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
This pull request fixes the issue where the device (
CPUorCUDA) is not set correctly when using the--deviceargument in Hydra-configured scripts likersl_rl/train.pyandskrl/train.py. The bug caused the scripts to always default tocuda:0, even whencpuor a specific CUDA device (e.g.,cuda:1) was selected.The fix adds the following line to ensure that the selected device is properly set in
env_cfgbefore initializing the environment withgym.make():Fixes #1012
Type of change
Screenshots
Before:
[INFO]: Starting the simulation. This may take a few seconds. Please wait....After:
Checklist
pre-commitchecks with./isaaclab.sh --formatconfig/extension.tomlfileCONTRIBUTORS.mdor my name already exists there