Giters

lucidrains / vit-pytorch

Implementation of Vision Transformer, a simple way to achieve SOTA in vision classification with only a single transformer encoder, in Pytorch

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Multi-GPU training of NaViT model

b5y opened this issue · comments

commented

Hello!

I have a question about multi-GPU training using NaViT model.

I am able to run the training on 1 GPU, but not on several.

Any suggestions or ideas how it's possible to use multiple GPUs for training this particular model?

FYI: DP doesn't work straightaway.

Thank you in advance.

commented

Closing as there were some data-related issues. Works well with DP.