[RLlib] Fix bug in SingleAgentEnvRunner: Calling sample() would always force-reset the vector env (even if episodes were not completed in a previous call).
#41168
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
sven1977
requested review from
avnishn,
ArturNiederfahrenhorst,
smorad,
maxpumperla and
kouroshHakha
as code owners
kouroshHakha
approved these changes
Nov 16, 2023
rllib/algorithms/ppo/ppo.py
Outdated
Comment on lines
442
to
450
| # TEST | ||
| #import pickle | ||
| #import torch | ||
| #import os | ||
| #with open("train_batch.pkl", "wb") as file: | ||
| # pickle.dump(train_batch, file) | ||
| #torch.save(self.workers.local_worker().policy_map["default_policy"].model.state_dict(), "model_weights.pth") | ||
| # END TEST | ||
|
|
There was a problem hiding this comment.
Sorry, will comb through changes and clean everything up before merging. Thanks for this catch!
rllib/algorithms/ppo/ppo.py
Outdated
Comment on lines
453
to
463
| # TEST | ||
| #import pickle | ||
| #import torch | ||
| #import os | ||
| #from ray.rllib.utils.test_utils import check | ||
| #rw_path = "/Users/sven/ray_results/PPO_2023-11-15_13-28-16/PPO_CartPoleDebug_74001_00000_0_2023-11-15_13-28-16" | ||
| #with open(os.path.join(rw_path, "train_batch.pkl"), "rb") as file: | ||
| # rw_train_batch = pickle.load(file) | ||
| #rw_state_dict = torch.load(os.path.join(rw_path, "model_weights.pth")) | ||
| #self.workers.local_worker().module.load_state_dict(rw_state_dict) | ||
| # END TEST |
rllib/algorithms/ppo/ppo.py
Outdated
Comment on lines
480
to
482
| #check(train_batch["advantages"], rw_train_batch["default_policy"]["advantages"], rtol=0.000001) | ||
| #check(train_batch["vf_preds"], rw_train_batch["default_policy"]["vf_preds"]) | ||
| #check(train_batch["value_targets"], rw_train_batch["default_policy"]["value_targets"]) |
…single_agent_env_runner_force_reset_bug
sven1977
changed the title
sample() repeatedly would always force-reset the vector env (even if episodes were not completed in a previous call).sample() would always force-reset the vector env (even if episodes were not completed in a previous call).
ujjawal-khare
pushed a commit
to ujjawal-khare-27/ray
that referenced
this pull request
Nov 29, 2023
…ays force-reset the vector env (even if episodes were not completed in a previous call). (ray-project#41168)
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.
Fix bug in SingleAgentEnvRunner: Calling
sample()repeatedly would always force-reset the vector env (even if episodes were not completed in a previous call).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.