Closed Ruiyuan-Zhang closed 1 year ago
Please try using fairseq to setup the distributed environment (even with 1 GPU) for MoE stuff: https://github.com/shumingma/fairseq/blob/moe/fairseq/distributed/utils.py#L246
Please try using fairseq to setup the distributed environment (even with 1 GPU) for MoE stuff: https://github.com/shumingma/fairseq/blob/moe/fairseq/distributed/utils.py#L246
I have apply pytorch-lightning
for my distributed training. Is them conflict?
Please try using fairseq to setup the distributed environment (even with 1 GPU) for MoE stuff: https://github.com/shumingma/fairseq/blob/moe/fairseq/distributed/utils.py#L246
I have apply
pytorch-lightning
for my distributed training. Is them conflict?
TBH, I'm not familiar with pytorch-ligntning
. However, besides the modeling, MoE needs additional efforts on the training backend, so there may be some conflicts here.
okk, thanks for your help~
Hi,
I want to replace
Transformer Encoder
withX-MoE Encoder
. Below is my configuration:I faced the below Error:
Thanks for your help~