ciyongch commented on a change in pull request #16141: [mkldnn-v1.0] Add
MKL-DNN Convolution
URL: https://github.com/apache/incubator-mxnet/pull/16141#discussion_r324465357
##########
File path: src/operator/nn/mkldnn/mkldnn_base-inl.h
##########
@@ -277,6 +277,19 @@ inline static mkldnn::memory::desc GetWeightDesc(const
NDArray &arr,
}
}
+inline static const std::vector<NDArray> GetMKLDNNInputArray(const
std::vector<NDArray> &inputs) {
+ std::vector<NDArray> ret;
+ ret.reserve(inputs.size());
+ for (const auto &in : inputs) {
+ if (in.IsView() && in.IsMKLDNNData()) {
+ ret.push_back(in.Reorder2Default());
+ } else {
+ ret.push_back(in);
Review comment:
Return `inputs` here to avoid overhead?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
