It transfers any tensor to the defined GPU: mat1.to(device) ... dummy_input to linear transformation layer print (linear_output.shape) # output>> PyTorch.
確定! 回上一頁