multi GPUs error: RuntimeError: arguments are located on different GPUs

[JingyunLiang]

For multi GPU, it outputs:

RuntimeError: arguments are located on different GPUs at /pytorch/torch/lib/THC/generic/THCTensorMathBlas.cu:236

How to fix it?

[dichen-cd]

Currently I don't have a multi-gpu environment, sorry about that. You are welcomed to create a PR if you have a solution.

Just by any chance it works, try reduce the batch/tensor size.

[mxtx0509]

Hello, is your problem solved? Can you train on multiple GPUs? Thanks

[mxtx0509]

I encountered the same problem during multi-GPU training.