projects / platform / upstream / tensorflow.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: ad07a86) | patch
[BatchNorm] Remove CPU implementation
authorYunxing Dai <yunxing@google.com>
Wed, 24 Jan 2018 19:42:22 +0000 (11:42 -0800)
committerTensorFlower Gardener <gardener@tensorflow.org>
Wed, 24 Jan 2018 19:45:50 +0000 (11:45 -0800)
commite8e1f1083d2c6ba4c2fb4e7e804d36624775e971
tree1a9ae985c5a958f72dbe66714fc9703987ae4258tree | snapshot
parentad07a86d75ab06bbcfd6f8f6a24debd9036a52d0commit | diff
[BatchNorm] Remove CPU implementation

We now use batchnorm rewriter (tensorflow/compiler/xla/service/batchnorm_rewriter.h) to expand batch norm into smaller ops. A specific implementation should not be needed anymore (for CPU).

RELNOTES:n/a
PiperOrigin-RevId: 183117252
tensorflow/compiler/xla/service/cpu/ir_emitter.cc diff | blob | history
tensorflow/compiler/xla/service/cpu/ir_emitter.h diff | blob | history
Domain: Machine Learning / ML Framework;