verl-project/verl

Fused CE loss integration

Open

#97 opened on Jan 12, 2025

View on GitHubContributor guide
 (3 comments) (1 reaction) (0 assignees)Python (4,410 forks)auto 404
call for contributionhelp wanted

Repository metrics

Stars
 (22,971 stars)
PR merge metrics
 (Avg merge 3d 6h) (125 merged PRs in 30d)

Description

Integrate it with main stream models: https://github.com/apple/ml-cross-entropy so that model with large vocab size uses much less memory

Contributor guide

Research direction
Explore the fusion of the cross entropy loss from apple/ml cross entropy into the training loop of main models in verl project/verl. Identify the model's current loss function and replace it with the fused version, ensuring gradient computation and memory savings.
Tech stack
pythonpytorch
Domain
backendai
Issue type
Feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
2
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.
Clear
Prerequisites
PythonPyTorch
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
60