MiniCPM-V
MiniCPM-V copied to clipboard
Published
3 months ago
•
OpenBMB
OpenBMB
[BUG] <title> 请问在用deepspeed zero3 训练的过程因为minicpm navit的逻辑会导致不同rank上的image feature size 不同这样会hang住，这个是怎么解决的？zero2是没有问题的
是否已有关于该错误的issue或讨论？ | Is there an existing issue / discussion for this?
- [X] 我已经搜索过已有的issues和讨论 | I have searched the existing issues / discussions
该问题是否在FAQ中有解答？ | Is there an existing answer for this in FAQ?
- [X] 我已经搜索过FAQ | I have searched FAQ
当前行为 | Current Behavior
期望能知道这里的解决方案
期望行为 | Expected Behavior
No response
复现方法 | Steps To Reproduce
No response
运行环境 | Environment
- OS:
- Python:
- Transformers:
- PyTorch:
- CUDA (`python -c 'import torch; print(torch.version.cuda)'`):
备注 | Anything else?
No response
这个大概率是还是deepspeed的版本和torch版本之类导致的，请问你用的是哪个版本的deepspeed，哪个版本的minicpmv呢
