when I set the "--gpu-id", default=2, I meet this problem. only "--gpu-id", default=0. I can train the model.

[zoufeng2416]

RuntimeError: Expected tensor for 'out' to have the same device as tensor for argument #3 'batch2'; but device 2 does not equal 0 (while checking arguments for baddbmm)

anyone could help me solve this problem?