[Feature] 请在pipeline parallel上支持不同长度的input, label的计算。
Open
enhancement
- Dominant language
- Python
- Stars
- 421
- Forks
- 67
- PR merge metrics
- No merged PRs in 30d
Description
### Describe the feature
在目前的情形下，pipeline parallel会在第一份数据进入时固定logits等的形状，但事实上，下游任务中很多时候最好能支持变长以节省算力，而不是进行过多的padding。
### Will you implement it?
- [ ] I would like to implement this feature and create a PR!
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.