[RLlib] Cleanup examples folder 18: Add example script for offline RL (BC) training on single-agent, while evaluating w/ multi-agent setup. #46251
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
…nup_examples_folder_18_bc_sa_training_then_ma_evaluation
sven1977
requested review from
ArturNiederfahrenhorst and
simonsays1980
as code owners
simonsays1980
approved these changes
Jun 25, 2024
| train_batch = train_batch.as_multi_agent() | ||
| self._counters[NUM_AGENT_STEPS_SAMPLED] += train_batch.agent_steps() | ||
| self._counters[NUM_ENV_STEPS_SAMPLED] += train_batch.env_steps() | ||
| # TODO (sven): Use metrics API as soon as we moved to new API stack |
There was a problem hiding this comment.
Is this one not using the MetricsLogger, yet? I use it in my overhaul of offline RL
There was a problem hiding this comment.
Yes, you should. All good there. :)
But the hybrid API stack still goes through the summarize_episodes utility inside algorithm.py, which operates on the old RolloutMetrics objects returned by RolloutWorkers.
| Here, SA=single-agent and MA=multi-agent. | ||
|
|
||
| Note that the BC Algorithm - by default - runs on the hybrid API stack, using RLModules, | ||
| but not EnvRunners or ConnectorV2s yet. |
There was a problem hiding this comment.
Probably adding here that is specifically doe snot use SingleAgentEpisode/MultiAgentEpisode?
| base_config = ( | ||
| BCConfig() | ||
| .environment( | ||
| observation_space=dummy_env.observation_space, |
There was a problem hiding this comment.
Can we give a quick note, why in this case the user needs to provide the spaces?
…nup_examples_folder_18_bc_sa_training_then_ma_evaluation
…nup_examples_folder_18_bc_sa_training_then_ma_evaluation
…nup_examples_folder_18_bc_sa_training_then_ma_evaluation
sven1977
deleted the
cleanup_examples_folder_18_bc_sa_training_then_ma_evaluation
branch
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.
Cleanup examples folder 18: Add example script for offline RL (BC) training on single-agent, while evaluating w/ multi-agent setup.
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.