diff options
| author | Mike Kelly <mike.kelly@arm.com> | 2022-04-21 11:57:09 +0100 |
|---|---|---|
| committer | mike.kelly <mike.kelly@arm.com> | 2022-05-05 08:29:20 +0000 |
| commit | 1299496996bc332f02218f926640a9255ed60310 (patch) | |
| tree | 2d242e142bd8fe7387140bcf8cdf39cd13ffc9eb /src/backends/neon/workloads/NeonUnidirectionalSequenceLstmFloatWorkload.hpp | |
| parent | 8272a7bda2974c39b6c06e3eb3a000f2bdb749f7 (diff) | |
| download | armnn-1299496996bc332f02218f926640a9255ed60310.tar.gz | |
IVGCVSW-6806 Add Unidirectional Sequence Lstm support to Neon
* Corrected TensorInfo order for IsUnidirectionalSequenceLstmSupported
* outputStateOut TensorInfo is not optional.
* cellStateOut TensorInfo is not optional.
* TensorInfo Order matches other QLSTM/LSTM layers.
* Added missing parameters to UnidirectionalSequenceLstmOperator for
delegate.
* Added quantized UnidirectionalSequenceLstm support to Neon
!android-nn-driver:7457
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: I26dde1bb96793dd25eb9081ca5ae5f63752288c4
Diffstat (limited to 'src/backends/neon/workloads/NeonUnidirectionalSequenceLstmFloatWorkload.hpp')
| -rw-r--r-- | src/backends/neon/workloads/NeonUnidirectionalSequenceLstmFloatWorkload.hpp | 5 |
1 files changed, 2 insertions, 3 deletions
diff --git a/src/backends/neon/workloads/NeonUnidirectionalSequenceLstmFloatWorkload.hpp b/src/backends/neon/workloads/NeonUnidirectionalSequenceLstmFloatWorkload.hpp index 776afd3965..48cf7dc7e4 100644 --- a/src/backends/neon/workloads/NeonUnidirectionalSequenceLstmFloatWorkload.hpp +++ b/src/backends/neon/workloads/NeonUnidirectionalSequenceLstmFloatWorkload.hpp @@ -10,7 +10,6 @@ #include <armnn/backends/Workload.hpp> #include <armnn/backends/WorkloadData.hpp> -#include "arm_compute/graph/Tensor.h" #include "arm_compute/runtime/NEON/functions/NELSTMLayer.h" #include "arm_compute/runtime/NEON/functions/NEPermute.h" #include "arm_compute/runtime/NEON/functions/NESplit.h" @@ -86,9 +85,9 @@ arm_compute::Status NeonUnidirectionalSequenceLstmFloatWorkloadValidate(const TensorInfo& input, const TensorInfo& outputStateIn, const TensorInfo& cellStateIn, + const TensorInfo& outputStateOut, + const TensorInfo& cellStateOut, const TensorInfo& output, - const Optional<TensorInfo>& hiddenStateOutput, - const Optional<TensorInfo>& cellStateOutput, const UnidirectionalSequenceLstmDescriptor& descriptor, const LstmInputParamsInfo& paramsInfo); |