Oleksiy Syvokon ee5de0ba44 BERT decoder: Fix causal mask dtype. ...

PyTorch < 1.3 requires multiplication operands to be of the same type.
This was violated when using default attention mask (i.e.,
attention_mask=None in arguments) given BERT in the decoder mode.

In particular, this was breaking Model2Model and made tutorial
from the quickstart failing.

2020-02-11 15:19:22 -05:00

20 KiB

Raw Blame History

View Raw