devtools/eigen
1
0
Fork 0
mirror of https://gitlab.com/libeigen/eigen.git synced 2026-04-10 11:34:33 +08:00
Code Issues Packages Projects Releases Wiki Activity

Updated the tensor sum and mean reducer to enable them to process complex numbers on cuda gpus.

Browse Source
This commit is contained in:
Benoit Steiner
2016-09-28 17:08:41 -07:00
parent f3a00dd2b5
commit 2bda1b0d93
2 changed files with 41 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
unsupported/Eigen/CXX11/src/Tensor/TensorFunctors.h
View File

@@ -99,7 +99,8 @@ template <typename T> struct SumReducer
static const bool IsStateful = false;
EIGEN_DEVICE_FUNC EIGEN_STRONG_INLINE void reduce(const T t, T* accum) const {
(*accum) += t;
internal::scalar_sum_op<T> sum_op;
*accum = sum_op(*accum, t);
}
template <typename Packet>
EIGEN_DEVICE_FUNC EIGEN_STRONG_INLINE void reducePacket(const Packet& p, Packet* accum) const {
@@ -145,7 +146,8 @@ template <typename T> struct MeanReducer
MeanReducer() : scalarCount_(0), packetCount_(0) { }
EIGEN_DEVICE_FUNC EIGEN_STRONG_INLINE void reduce(const T t, T* accum) {
(*accum) += t;
internal::scalar_sum_op<T> sum_op;
*accum = sum_op(*accum, t);
scalarCount_++;
}
template <typename Packet>