onnxruntime/orttraining/orttraining
liqunfu fe50213491
Liqun/bert pretrain2 (#5327)
* bert single node multi GPU pretrain w/o checkpoint

Co-authored-by: liqun <liqun@OrtTrainingDev4.af05slrtruoetgaxwwjv5nsq5e.px.internal.cloudapp.net>
2020-10-01 11:01:26 -07:00
..
core Enable BiasDropoutFusion for CUDA EP only (#5324) 2020-09-29 14:00:15 -07:00
models Transformer layer-wise Recompute (#4526) 2020-09-24 19:56:32 -07:00
python Liqun/bert pretrain2 (#5327) 2020-10-01 11:01:26 -07:00
test Liqun/bert pretrain2 (#5327) 2020-10-01 11:01:26 -07:00
training_ops Scale Op for ReduceMeanGrad. (#5191) 2020-09-29 09:30:49 +08:00