Transformer/Transformer(Greedy_decoder)-Torch.py on gpu

Hello, I want to put the Transformer (Greedy_decoder)-Torch.py code on the gpu, using model=model.to(device), input_data also to (device), but the error still appears "Expected object of backend CUDA but backend CPU for argument #2 'mat2”

Start with Transformer/Transformer(Greedy_decoder)-Torch.py and reproduce the reported CUDA-versus-CPU mat2 error. Trace the devices of model, input_data, and the tensors involved in the failing operation. Done means the example runs on the GPU without a backend mismatch.

Written by the indexing model from the issue text.