diff options
| author | Giorgio Arena <giorgio.arena@arm.com> | 2018-07-06 17:06:36 +0100 |
|---|---|---|
| committer | Anthony Barbier <anthony.barbier@arm.com> | 2018-11-02 16:54:54 +0000 |
| commit | 6200fa405b16b4145b926a96de197718ad31bf93 (patch) | |
| tree | 4ecaa3a29d79371c6439acf5bb580bc7ba99af09 /arm_compute | |
| parent | ea55f91e5dd4e5bc766fabbac6df6ce3ab984d0e (diff) | |
| download | ComputeLibrary-6200fa405b16b4145b926a96de197718ad31bf93.tar.gz | |
COMPMID-1288 Optimizing CLGEMMLowp using 8 bit dot product instruction
Change-Id: I536174b9381660a94578d6aa1892a6289a820391
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/139109
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'arm_compute')
| -rw-r--r-- | arm_compute/core/CL/CLKernelLibrary.h | 6 | ||||
| -rw-r--r-- | arm_compute/runtime/NEON/functions/NEFullyConnectedLayer.h | 2 |
2 files changed, 7 insertions, 1 deletions
diff --git a/arm_compute/core/CL/CLKernelLibrary.h b/arm_compute/core/CL/CLKernelLibrary.h index f25915f053..18b6bb4434 100644 --- a/arm_compute/core/CL/CLKernelLibrary.h +++ b/arm_compute/core/CL/CLKernelLibrary.h @@ -277,6 +277,12 @@ public: return _context; } + /** Gets the CL device for which the programs are created. */ + cl::Device &get_device() + { + return _device; + } + /** Sets the CL device for which the programs are created. * * @param[in] device A CL device. diff --git a/arm_compute/runtime/NEON/functions/NEFullyConnectedLayer.h b/arm_compute/runtime/NEON/functions/NEFullyConnectedLayer.h index d4166b3830..33ac8ecb8a 100644 --- a/arm_compute/runtime/NEON/functions/NEFullyConnectedLayer.h +++ b/arm_compute/runtime/NEON/functions/NEFullyConnectedLayer.h @@ -57,7 +57,7 @@ public: * @param[in] is_batched_fc_layer True if it is a batched fully connected layer */ void configure(const ITensor *input, ITensor *output, bool transpose_weights, bool is_batched_fc_layer); - /** Static function to check if given info will lead to a valid configuration of @ref CLFullyConnectedLayerReshapeWeights + /** Static function to check if given info will lead to a valid configuration of @ref NEFullyConnectedLayerReshapeWeights * * @param[in] input Weights tensor info. The weights must be 2 dimensional. Data types supported: F32. * @param[in] output Destination tensor info. Data type supported: Same as @p input. |