Closed VProv closed 3 years ago
Dear Rico,
I have been approached several times by people who have learned models by applying BPE-Dropout only once on the training corpus. I think this note can save many GPU-hours in the future.
Ivan
thanks!
Dear Rico,
I have been approached several times by people who have learned models by applying BPE-Dropout only once on the training corpus. I think this note can save many GPU-hours in the future.
Ivan