Lightning-AI/pytorch-lightning

Provide a non-expert mode for the DeviceStatsMonitor

Open

#18,652 创建于 2023年9月27日

在 GitHub 查看
 (0 评论) (15 反应) (0 负责人)Python (3,233 fork)batch import
callback: device statsfeaturehelp wanted

仓库指标

Star
 (26,687 star)
PR 合并指标
 (平均合并 9天 15小时) (30 天内合并 3 个 PR)

描述

Description & Motivation

The DeviceStatsMonitor currently dumps the entire torch.cuda.memory_stats into the logger. Beginner users might be overwhelmed by the amount of data logged and are confused which metrics matter to maximize their utilization.

Pitch

Make the callback more user-friendly by logging only core metrics like GPU utilization and total memory usage. Provide an expert mode that when enabled logs what we see today.

Alternatives

No response

Additional context

This was proposed by user magical_arjay on our Discord.

cc @borda @awaelchli

贡献者指南