verl-project/verl

[FR] Better metric reduction utility

Open

#1,508 opened on May 13, 2025

View on GitHubContributor guide
 (5 comments) (2 reactions) (0 assignees)Python (4,413 forks)auto 404
enhancementfeature requestgood first issue

Repository metrics

Stars
 (22,985 stars)
PR merge metrics
 (Avg merge 3d 6h) (125 merged PRs in 30d)

Description

Motivation

reduce_metrics seems to be used for reducing metrics across DP ranks and batches, which only simply means the values.

However, reducing metrics across DP ranks is non-trivial since some metrics like "mean" are not linear.

Contributor guide

Research direction
Investigate the current reduce metrics implementation in verl/utils/metric/utils.py and determine how to properly reduce non linear metrics like mean across DP ranks. Consider using atomic reductions or computing sum and count separately.
Tech stack
python
Domain
aimachine learningbackend
Issue type
Feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
Half day
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Mostly clear
Prerequisites
Python
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
60