diff options
author | Gian Marco Iodice <gianmarco.iodice@arm.com> | 2017-06-30 12:21:00 +0100 |
---|---|---|
committer | Anthony Barbier <anthony.barbier@arm.com> | 2018-09-17 14:15:39 +0100 |
commit | bdb6b0bb156588dc39fd5084d4c91d05b5148610 (patch) | |
tree | bb3c3645dd9abbf20462dace7828bb7ec459dc4d /arm_compute/core/NEON/kernels/NEGEMMMatrixAdditionKernel.h | |
parent | ac69aa137e360340fe9f148f019d93af6c3d8336 (diff) | |
download | ComputeLibrary-bdb6b0bb156588dc39fd5084d4c91d05b5148610.tar.gz |
COMPMID-433 - Port NEGEMM to support 16 bit fixed point
Change-Id: I82de74d7027bbc8a00a4d6671e968785280d5f6c
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/79498
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Moritz Pflanzer <moritz.pflanzer@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Diffstat (limited to 'arm_compute/core/NEON/kernels/NEGEMMMatrixAdditionKernel.h')
-rw-r--r-- | arm_compute/core/NEON/kernels/NEGEMMMatrixAdditionKernel.h | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/arm_compute/core/NEON/kernels/NEGEMMMatrixAdditionKernel.h b/arm_compute/core/NEON/kernels/NEGEMMMatrixAdditionKernel.h index 1ab52fa2f2..5cdcc95ee9 100644 --- a/arm_compute/core/NEON/kernels/NEGEMMMatrixAdditionKernel.h +++ b/arm_compute/core/NEON/kernels/NEGEMMMatrixAdditionKernel.h @@ -55,7 +55,7 @@ public: * * @note The input and output tensor must have the same dimensions * - * @param[in] input Input tensor (Matrix C). Data types supported: QS8/F16/F32 + * @param[in] input Input tensor (Matrix C). Data types supported: QS8/QS16/F16/F32 * @param[in, out] output Output tensor. If this kernel is used to finalize the GEMM result, output contains the result obtained by the kernel @ref NEGEMMMatrixMultiplyKernel. Data type supported: the same as @p input. * @param[in] beta Weight of matrix C */ @@ -67,7 +67,7 @@ public: private: /** Common signature for all the matrix addition functions * - * @param[in] input An input tensor. Data types supported: QS8/F16/F32 + * @param[in] input An input tensor. Data types supported: QS8/QS16/F16/F32 * @param[out] output The output tensor. Data type supported: same as @p input * @param[in] window Region on which to execute the kernel. * @param[in] beta Weight of matrix C |