Suffian Khan 498f94668d Keep all_finite tensor on CPU when using PyTorch Frontend (#5371)		2020-10-08 15:47:18 -07:00
..
core	rename all TransposeMatMul nodes to FusedMatMul (#5373)	2020-10-05 12:41:05 -07:00
models	Adding an option for cudnn conv algorithms. (#5159)	2020-10-05 16:53:52 -07:00
python	Keep all_finite tensor on CPU when using PyTorch Frontend (#5371)	2020-10-08 15:47:18 -07:00
test	Liqun/bert pretrain tb (#5377)	2020-10-06 16:28:31 -07:00
training_ops	Scale Op for ReduceMeanGrad. (#5191)	2020-09-29 09:30:49 +08:00