xnn.common.train.trainer.resolve_device#

xnn.common.train.trainer.resolve_device(name)[source]#

Resolve a device specification string into a torch.device.

Parameters:: name (str) – Device name. The special value "auto" selects "cuda" when a CUDA device is available and falls back to "cpu" otherwise. Any other value is passed through to torch.device unchanged (e.g. "cpu", "cuda", "cuda:1").
Returns:: The resolved device.
Return type:: torch.device