最新版的master编译的代码分布式仍然无法evaluate和save embedding
- Dominant language
- C++
- Stars
- 2.9k
- Forks
- 553
- PR merge metrics
- No merged PRs in 30d
Description
运行的是示例脚本run_dist_euler.sh
节点0日志：
节点1日志：
二者无限输出1 workers have finished。。。
请教下是什么原因导致的呢？SyncExitHook方法中_num_finished_workers不同worker之间是共享的吗？感觉是这个出的问题。谢谢指导！
Contributor guide
No contributing guide indexed for this repository
Research direction
Start by running the example script run_dist_euler.sh and comparing the two worker logs. Read SyncExitHook, especially how _num_finished_workers is handled across workers; done means distributed evaluation and embedding saving complete without repeatedly printing that one worker has finished.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- distributed-systems, machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100