[人工智能]RuntimeError: module must have its parameters and buffers on device cuda:0 (device_ids[0]) but found

问题

RuntimeError: module must have its parameters and buffers on device cuda:0 (device_ids[0]) but found one of them on device: cuda:1
RuntimeError: module必须在设备cuda:0 (device_ids[0])上有它的参数和缓冲区，但在设备:cuda:1上找到其中一个

解决：

将

model = torch.nn.DataParallel(model)

改为

model = torch.nn.DataParallel(model, device_ids=[1])

或

model = torch.nn.DataParallel(model, device_ids=[1,2])