pufferl: log obs max/min/mean per eval step#458
Merged
Conversation
Three new wandb keys (`obs/max`, `obs/min`, `obs/mean`) collected every env step in evaluate() and rolled up by mean_and_log. Each value is the scalar reduction across the full batch and obs-dim, so the wandb panel shows the absolute range of every feature the policy sees over the eval window. Useful for catching clipping, mis-normalization, or unbounded features post-training-tweak. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
There was a problem hiding this comment.
Pull request overview
Adds three per-env-step observation statistics (obs/max, obs/min, obs/mean) to PuffeRL.evaluate() so they are surfaced in W&B (as environment/obs/{max,min,mean}) to help catch clipping/normalization regressions.
Changes:
- Append
o_device.max/min/mean().item()intoself.statson each evaluate step, before the policy forward pass.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
vcharraut
approved these changes
May 30, 2026
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Adds three wandb scalars — `obs/max`, `obs/min`, `obs/mean` — computed each env step in `PuffeRL.evaluate()` from the same tensor that's about to enter the policy