Brutex
/
cavis
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

no openmp for ClipByGlobalNorm 

Signed-off-by: raver119 <raver119@gmail.com>
master
raver119 2019-08-15 18:24:29 +03:00
parent 935cbf5df3
commit 72cb5936f1
1 changed files with 0 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
libnd4j/include/ops/declarable/helpers/cuda/transforms.cu
View File

@ -806,7 +806,6 @@ void clipByNormBP(nd4j::LaunchContext* context, const NDArray& input, const NDAr
void clipByGlobalNorm_(nd4j::LaunchContext * context, std::vector<NDArray*> const& inputs, double clipNorm, nd4j::memory::Workspace* workspace, std::vector<NDArray*>& outputs, bool isInplace) {
NDArray globalNorm = NDArrayFactory::create<T>(0, inputs[0]->getContext()); //sqrt(sum([l2norm(t)**2 for t in t_list]))
PRAGMA_OMP_PARALLEL_FOR
for (auto i = 0; i < inputs.size(); i++) {
auto input = inputs[i];
auto l2norm = input->reduceNumber(reduce::Norm2);
@ -818,7 +817,6 @@ void clipByNormBP(nd4j::LaunchContext* context, const NDArray& input, const NDAr
globalNorm.syncToHost();
const T factor = clipNorm / globalNorm.e<T>(0);
PRAGMA_OMP_PARALLEL_FOR
for (size_t e = 0; e < inputs.size(); e++) {
// all-reduce
auto input = inputs[e];