diff options
author | Narumol Prangnawarat <narumol.prangnawarat@arm.com> | 2019-05-20 15:31:05 +0100 |
---|---|---|
committer | Matteo Martincigh <matteo.martincigh@arm.com> | 2019-05-23 13:37:29 +0000 |
commit | 15eb5832f45d35c5041ba35a43787e8003e22edb (patch) | |
tree | 09fed880bfb9f384d3170aad5c76e4d565267e20 /src/backends/cl/ClLayerSupport.hpp | |
parent | 495852f2adef1d11fbf13ce6347cf61973ce1a65 (diff) | |
download | armnn-15eb5832f45d35c5041ba35a43787e8003e22edb.tar.gz |
IVGCVSW-2771 Fix SubTensor error in vgg16 ExecuteNetwork NEON
* Add check if Sub-tensors cannot be used, call ACL function
* Add computation of SplitAxis from SplitterDescriptor
* Add NeonSplitterWorkload functions
* Modify IsSplitterSupported to call ACL validate function
if sub-tensor cannot be used
* Also check if quantization parameters match when using sub-tensors
* Add more unit tests for Splitter in TfParser and TfLiteParser
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: I31e4c7d055117c83c65b598c4125442173242226
Diffstat (limited to 'src/backends/cl/ClLayerSupport.hpp')
-rw-r--r-- | src/backends/cl/ClLayerSupport.hpp | 6 |
1 files changed, 6 insertions, 0 deletions
diff --git a/src/backends/cl/ClLayerSupport.hpp b/src/backends/cl/ClLayerSupport.hpp index b634d46768..fca0bfd352 100644 --- a/src/backends/cl/ClLayerSupport.hpp +++ b/src/backends/cl/ClLayerSupport.hpp @@ -200,10 +200,16 @@ public: const SpaceToBatchNdDescriptor& descriptor, Optional<std::string&> reasonIfUnsupported = EmptyOptional()) const override; + ARMNN_DEPRECATED_MSG("Use IsSplitterSupported with outputs instead") bool IsSplitterSupported(const TensorInfo& input, const ViewsDescriptor& descriptor, Optional<std::string&> reasonIfUnsupported = EmptyOptional()) const override; + bool IsSplitterSupported(const TensorInfo& input, + const std::vector<std::reference_wrapper<TensorInfo>>& outputs, + const ViewsDescriptor& descriptor, + Optional<std::string&> reasonIfUnsupported = EmptyOptional()) const override; + bool IsStridedSliceSupported(const TensorInfo& input, const TensorInfo& output, const StridedSliceDescriptor& descriptor, |