出现上述错误,是因为DP加载,默认会将模型和参数放在GPU设备上,如果想用cpu去推理DP加载的模型,必须对DP 模型进行下面一步转换才可以.
将torch.nn.DataParalle 放到cpu上
model = torch.nn.DataParallel(model, device_ids=[0,1,2,3])
model = model.module.to(torch.device('cpu'))文章出处登录后可见！
已经登录？立即刷新
module must have its parameters and buffers on device cuda:0 (device_ids[0]) but found one of them
赞 (1)
