diff options
author | Xiaoqiang Zheng <zhengxq@google.com> | 2016-05-07 22:22:44 -0800 |
---|---|---|
committer | TensorFlower Gardener <gardener@tensorflow.org> | 2016-05-09 09:21:14 -0700 |
commit | 6ff265ebae14586db5db623b7502ddbdbc8bbd12 (patch) | |
tree | d4df06ea887dfb603b22c9df1147b41b38994b54 /tensorflow/stream_executor/dnn.cc | |
parent | a846576a67da4bdd0b610a17b5c8d0d92e41f094 (diff) |
Adding autotune to the Cudnn conv algorithm selection.
For now, use TF_CUDNN_USE_AUTOTUNE=1 to enable this feature. Once it is mature
enough, it will be turned on by default.
Support for the backward steps will be added later.
Change: 121769364
Diffstat (limited to 'tensorflow/stream_executor/dnn.cc')
-rw-r--r-- | tensorflow/stream_executor/dnn.cc | 5 |
1 files changed, 5 insertions, 0 deletions
diff --git a/tensorflow/stream_executor/dnn.cc b/tensorflow/stream_executor/dnn.cc index 41f3bf83d3..7e98db723f 100644 --- a/tensorflow/stream_executor/dnn.cc +++ b/tensorflow/stream_executor/dnn.cc @@ -22,6 +22,11 @@ namespace perftools { namespace gputools { namespace dnn { +bool DnnSupport::GetConvolveAlgorithms( + std::vector<AlgorithmType>* out_algorithms) { + return false; +} + string QuantizedActivationModeString(QuantizedActivationMode mode) { switch (mode) { case dnn::QuantizedActivationMode::k8Bit: |