AlibabaResearch / AlibabaResearch/AdvancedLiterateMachinery
采用推荐脚本训练，在3万步左右会出现Loss is nan, stopping training，又遇到的嘛？
Open
- Dominant language
- C++
- Stars
- 1.8k
- Forks
- 195
- PR merge metrics
- No merged PRs in 30d
Description
Loss is nan, stopping training
{'pt_loss': tensor(nan, device='cuda:0', grad_fn=), 'poly_loss': tensor(nan, device='cuda:0', grad_fn=), 'rec_loss': tensor(3.2648, device='cuda:0', grad_fn=)}
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.