You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
DIstilBert model donesn't support f16 or mixed precision training.
The hard-coded uses of float32 from the tensorflow implementation of BERT and ELECTRA has been fixed by #3320 .
The same problem in DistilBert needs a fix.
The text was updated successfully, but these errors were encountered:
🚀 Feature request
DIstilBert model donesn't support f16 or mixed precision training.
The hard-coded uses of float32 from the tensorflow implementation of BERT and ELECTRA has been fixed by #3320 .
The same problem in DistilBert needs a fix.
The text was updated successfully, but these errors were encountered: