Xiaomeng Yang 2ce39de3fc Add elementwise_affine for layer_norm_op (#19713) ...

Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/19713

Add elementwise_affine for layer_norm_op

Reviewed By: houseroad

Differential Revision: D15075454

fbshipit-source-id: e8a7d3da1c81e49fa55323f5e74a68bc4ef8d83f

2019-04-26 17:20:01 -07:00

..

broadcast.cc

Separate elementwise level2 math functions (#16753)

2019-02-07 18:38:26 -08:00

broadcast.cu

Separate elementwise level2 math functions (#16753)

2019-02-07 18:38:26 -08:00

broadcast.h

Separate elementwise level2 math functions (#16753)

2019-02-07 18:38:26 -08:00

elementwise.cc

Add gelu op (#18992)

2019-04-08 21:58:29 -07:00

elementwise.cu

Add gelu op (#18992)

2019-04-08 21:58:29 -07:00

elementwise.h

Add gelu op (#18992)

2019-04-08 21:58:29 -07:00

half_utils.h

Separate elementwise level2 math functions (#16753)

2019-02-07 18:38:26 -08:00

reduce.cc

Add elementwise_affine for layer_norm_op (#19713)

2019-04-26 17:20:01 -07:00

reduce.cu

Add elementwise_affine for layer_norm_op (#19713)

2019-04-26 17:20:01 -07:00

reduce.cuh

Separate gpu reduce functions (#17146)

2019-02-20 14:49:01 -08:00

reduce.h

Separate reduce functions from math (#16929)

2019-02-13 17:50:47 -08:00

transpose.cc

Update math::Transpose to support tensor with size > 2G (#17670)

2019-03-20 18:22:21 -07:00

transpose.cu

Update math::Transpose to support tensor with size > 2G (#17670)

2019-03-20 18:22:21 -07:00

transpose.h

Update math::Transpose to support tensor with size > 2G (#17670)

2019-03-20 18:22:21 -07:00

utils.cc

optimize BatchMatmulOp (#18612)

2019-04-23 15:34:59 -07:00

utils.h

optimize BatchMatmulOp (#18612)

2019-04-23 15:34:59 -07:00