apeforest commented on a change in pull request #16218: Improving performance
of argmax operator
URL: https://github.com/apache/incubator-mxnet/pull/16218#discussion_r326474610
##########
File path: src/operator/tensor/broadcast_reduce_op.h
##########
@@ -556,6 +556,162 @@ inline bool ReduceAxesOpForwardStorage(const
nnvm::NodeAttrs& attrs,
return dispatched;
}
+struct argmax {
+ template<typename DType>
+ MSHADOW_XINLINE static void Map(int i, const int nWorkers, const DType
*in_data, DType *out_data,
Review comment:
Please use `index_t` for Map index so that we can support large tensor.
----------------------------------------------------------------
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
