Closed jasonlyik closed 10 months ago
PyTorch recurrent layers (LSTM/GRU) have two bias vectors for CuDNN compatibility.
In footprint calculation the second bias vector b_hh should be skipped or suppressed
b_hh
This issue potentially affects the LSTMCell used in the MG task
Will not fix this issue since it is characteristic of the particular model, which is what the footprint metric captures.
PyTorch recurrent layers (LSTM/GRU) have two bias vectors for CuDNN compatibility.
In footprint calculation the second bias vector
b_hh
should be skipped or suppressed