[RLlib] Use config (not self.config) in Learner.compute_loss_for_module to prepare these for multi-agent-capability.
#45053
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
Signed-off-by: sven1977 <[email protected]>
sven1977
requested review from
avnishn,
ArturNiederfahrenhorst,
maxpumperla,
kouroshHakha and
simonsays1980
as code owners
…ics_do_over_03_learner_on_new_metrics_logger Signed-off-by: sven1977 <[email protected]> # Conflicts: # rllib/algorithms/ppo/ppo_learner.py # rllib/algorithms/ppo/tf/ppo_tf_learner.py # rllib/algorithms/ppo/torch/ppo_torch_learner.py
Signed-off-by: sven1977 <[email protected]>
sven1977
added
the
tests-ok
simonsays1980
approved these changes
May 2, 2024
| @@ -72,7 +72,7 @@ def compute_loss_for_module( | |||
| trajectory_len=rollout_frag_or_episode_len, | |||
| recurrent_seq_len=recurrent_seq_len, | |||
| ) | |||
| if self.config.enable_env_runner_and_connector_v2: | |||
| if config.enable_env_runner_and_connector_v2: | |||
There was a problem hiding this comment.
I wonder, if the new env runners work on APPO/IMPALA. In my test case they do not in the MA case where a list of episodes is tried to be compressed_if_needed
There was a problem hiding this comment.
IMPALA and APPO are WIP on the new EnvRunners, officially not supported yet.
https://docs.ray.io/en/master/rllib/rllib-new-api-stack.html
| @@ -61,7 +61,7 @@ def compute_loss_for_module( | |||
| ).squeeze() | |||
|
|
|||
| # Use double Q learning. | |||
| if self.config.double_q: | |||
| if config.double_q: | |||
There was a problem hiding this comment.
Great catch. I had the same changed in another PR on the side. This would have been led to some bugs in MA off-policy.
| ).items() | ||
| } | ||
| ) | ||
| for module_id in set(loss_per_module.keys()) - {ALL_MODULES}: |
…self_config_in_learner_compute_loss Signed-off-by: sven1977 <[email protected]> # Conflicts: # rllib/algorithms/sac/torch/sac_torch_learner.py
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.
Use
config(notself.config) inLearner.compute_loss_for_moduleto prepare these for multi-agent-capability.compute_gradientsto NOT use DEFAULT_POLICY anymore, but to actually loop through the different RLModules.Why are these changes needed?
Related issue number
Checks
git commit -s) in this PR.scripts/format.shto lint the changes in this PR.method in Tune, I've added it in
doc/source/tune/api/under thecorresponding
.rstfile.