ZhennanQin commented on a change in pull request #14608: [MKLDNN]Refactor
requantize to speed up execution
URL: https://github.com/apache/incubator-mxnet/pull/14608#discussion_r279173400
##########
File path: src/operator/quantization/mkldnn/mkldnn_requantize-inl.h
##########
@@ -99,55 +108,47 @@ static void MKLDNNRequantizeForward(const
nnvm::NodeAttrs& attrs,
const std::vector<NDArray>& outputs) {
using namespace mshadow;
using namespace mxnet_op;
+ using red::limits::MaxValue;
+ using red::limits::MinValue;
typedef int32_t SrcDType;
typedef int8_t DstDType;
Review comment:
It could. But it's not introduced by this PR, so it doesn't necessary to
remove it in this PR.
----------------------------------------------------------------
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