Conversation
|
Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA). View this failed invocation of the CLA check for more information. For the most up to date status, view the checks section at the bottom of the pull request. |
Previously, EpisodeWrapper.step only summed rewards across action_repeat sub-steps, but state.metrics only contained the metrics from the last sub-step. This caused sparse or per-step metrics (e.g., action-change penalties) to only reflect the last sub-step's value. Fix: accumulate both rewards and metrics in the lax.scan loop, sum both across sub-steps, and write the summed metrics back to state.metrics. This matches the existing behavior for rewards and ensures all metrics are correctly aggregated. Fixes google#610
dafahaha
force-pushed
the
fix-episode-metrics
branch
from
September 17, 2026 01:10
1294d3e to
7f9bf49
Compare
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Fixes #610.
Previously,
EpisodeWrapper.steponly summed rewards acrossaction_repeatsub-steps, butstate.metricsonly contained the metrics from the last sub-step. This caused sparse or per-step metrics (e.g., action-change penalties) to only reflect the last sub-step's value, while earlier sub-step metrics were silently dropped.Changes
lax.scanloop inEpisodeWrapper.stepto accumulate both rewards and metricsaction_repeatsub-steps usingjax.tree.mapstate.metricsbefore aggregating intoepisode_metricsThis matches the existing behavior for rewards and ensures all metrics are correctly aggregated across sub-steps.
Testing
jax.tree.mapcorrectly handles the metrics dictionary structure