- 28 Aug, 2019 36 commits
-
-
Thomas Wolf authored
swap order of optimizer.step() and scheduler.step()
-
Thomas Wolf authored
distilbert: fix number of hidden_size
-
LysandreJik authored
-
Andreas Daiminger authored
-
Stefan Schweter authored
-
Thomas Wolf authored
Generative finetuning
-
Thomas Wolf authored
DilBERT
-
thomwolf authored
-
thomwolf authored
-
LysandreJik authored
-
LysandreJik authored
-
LysandreJik authored
-
thomwolf authored
-
thomwolf authored
-
thomwolf authored
-
thomwolf authored
-
VictorSanh authored
-
VictorSanh authored
-
VictorSanh authored
-
VictorSanh authored
-
VictorSanh authored
-
VictorSanh authored
-
VictorSanh authored
-
VictorSanh authored
-
VictorSanh authored
-
LysandreJik authored
-
LysandreJik authored
-
VictorSanh authored
-
VictorSanh authored
-
VictorSanh authored
-
VictorSanh authored
-
VictorSanh authored
-
VictorSanh authored
-
VictorSanh authored
-
VictorSanh authored
-
VictorSanh authored
-
- 27 Aug, 2019 4 commits
-
-
VictorSanh authored
-
VictorSanh authored
-
VictorSanh authored
-
Thomas Wolf authored
Change attention mask dtype to be bool. Fix #1119
-