[RLlib] Issue 18812: Torch multi-GPU stats not protected against race conditions. #18937
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
|
@mvindiola1 ^^ |
2 tasks
…e_18812_torch_multi_gpu_stats_race_condition
…e_18812_torch_multi_gpu_stats_race_condition
…e_18812_torch_multi_gpu_stats_race_condition # Conflicts: # rllib/agents/dqn/r2d2_torch_policy.py # rllib/agents/sac/rnnsac_torch_policy.py
| @@ -279,7 +291,7 @@ def extra_action_out_fn(policy: Policy, input_dict, state_batches, model, | |||
| postprocess_fn=postprocess_nstep_and_prio, | |||
| optimizer_fn=adam_optimizer, | |||
| extra_grad_process_fn=grad_process_and_td_error_fn, | |||
| extra_learn_fetches_fn=lambda policy: {"td_error": policy._td_error}, | |||
| extra_learn_fetches_fn=concat_multi_gpu_td_errors, | |||
There was a problem hiding this comment.
Might be better just to hardcode the lambda function directly as a function in the r2d2 policy class
| @@ -16,7 +16,7 @@ | |||
| from ray.rllib.policy.torch_policy import TorchPolicy | |||
| from ray.rllib.utils.annotations import override | |||
| from ray.rllib.utils.framework import try_import_torch | |||
| from ray.rllib.utils.torch_ops import huber_loss | |||
| from ray.rllib.utils.torch_ops import concat_multi_gpu_td_errors, huber_loss | |||
There was a problem hiding this comment.
Think its better not to abstract td_errors away
michaelzhiluo
approved these changes
Sep 30, 2021
…e_18812_torch_multi_gpu_stats_race_condition # Conflicts: # rllib/agents/impala/vtrace_torch_policy.py # rllib/policy/tf_policy_template.py # rllib/policy/torch_policy.py
…e_18812_torch_multi_gpu_stats_race_condition
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Outdated suggestions cannot be applied.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.
Issue 18812: Torch multi-GPU stats not protected against race conditions.
This PR:
Why are these changes needed?
#18812
Related issue number
Closes #18812
Checks
scripts/format.shto lint the changes in this PR.