projects / platform / upstream / tensorflow.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 22cb902) | patch
[XLA] Parallelize HloEvaluator::HandleConvolution
authorMichael Kuperstein <mkuper@google.com>
Sun, 8 Apr 2018 22:37:26 +0000 (15:37 -0700)
committerTensorFlower Gardener <gardener@tensorflow.org>
Sun, 8 Apr 2018 22:39:50 +0000 (15:39 -0700)
commit8fd805fc79ca585fe90ec9fd7c9e0feef89f798e
treeed1f57902eee9ed8b785168967c2e4463790e1a7tree | snapshot
parent22cb902924b3b6e243af6ee386e660c826d60d12commit | diff
[XLA] Parallelize HloEvaluator::HandleConvolution

This adds a parallel version of Literal::Populate, and uses it in the embarrassingly parallel convolution computation.

PiperOrigin-RevId: 192065277
tensorflow/compiler/xla/literal_util.h diff | blob | history
tensorflow/compiler/xla/literal_util_test.cc diff | blob | history
tensorflow/compiler/xla/service/hlo_evaluator.cc diff | blob | history
tensorflow/compiler/xla/shape_util.h diff | blob | history
tensorflow/compiler/xla/shape_util_test.cc diff | blob | history
Domain: Machine Learning / ML Framework;