diff options
| author | Georgios Pinitas <georgios.pinitas@arm.com> | 2020-04-29 11:44:10 +0100 |
|---|---|---|
| committer | Georgios Pinitas <georgios.pinitas@arm.com> | 2020-05-13 09:33:35 +0000 |
| commit | 5ce897f80a1a6ade8a07d61c7aaaf70d2aa5ee02 (patch) | |
| tree | dd87feb17198db35a5a838b6f9c14825ce25d02f /arm_compute/runtime/NEON/functions/NEWinogradConvolutionLayer.h | |
| parent | a1b8babbb492fa4cd3b392f6376a2dfa85fc854d (diff) | |
| download | ComputeLibrary-5ce897f80a1a6ade8a07d61c7aaaf70d2aa5ee02.tar.gz | |
COMPMID-3108: Add Winograd 3x3,4x4 FP16 support for NEON
Change-Id: I20680dc74a3d709297539e2132417308a7aecc9d
Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3159
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'arm_compute/runtime/NEON/functions/NEWinogradConvolutionLayer.h')
| -rw-r--r-- | arm_compute/runtime/NEON/functions/NEWinogradConvolutionLayer.h | 10 |
1 files changed, 6 insertions, 4 deletions
diff --git a/arm_compute/runtime/NEON/functions/NEWinogradConvolutionLayer.h b/arm_compute/runtime/NEON/functions/NEWinogradConvolutionLayer.h index 31047e8994..4a8fe61614 100644 --- a/arm_compute/runtime/NEON/functions/NEWinogradConvolutionLayer.h +++ b/arm_compute/runtime/NEON/functions/NEWinogradConvolutionLayer.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2017-2019 ARM Limited. + * Copyright (c) 2017-2020 ARM Limited. * * SPDX-License-Identifier: MIT * @@ -39,7 +39,9 @@ namespace arm_compute { +// Forward declarations class ITensor; + /** Basic function to simulate a convolution layer. This function calls the following NEON kernels: * -# @ref NEWinogradLayerTransformWeightsKernel (executed only once in the first call to the run() method ) * -# @ref NEWinogradLayerTransformInputKernel @@ -59,7 +61,7 @@ public: * * @param[in] input Source tensor. 3 lower dimensions represent a single input [width, height, IFM], * while every optional dimension from 4 and above represent a batch of inputs. - * Data types supported: F32. + * Data types supported: F16/F32. * @param[in] weights Weights tensor. Weights are 4D tensor with dimensions [kernel_x, kernel_y, IFM, OFM]. Data type supported: Same as @p input. * Currently only 3x3 and 5x5 kernels are supported. * @param[in] biases Biases tensor. Shared biases supported. Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p weights. @@ -81,7 +83,7 @@ public: * * @param[in] input Source tensor. 3 lower dimensions represent a single input [width, height, IFM], * while every optional dimension from 4 and above represent a batch of inputs. - * Data types supported: F32. + * Data types supported: F16/F32. * @param[in] weights Weights tensor. Weights are 4D tensor with dimensions [kernel_x, kernel_y, IFM, OFM]. Data type supported:Same as @p input. * Currently only 3x3 and 5x5 kernels are supported. * @param[in] biases Biases tensor. Shared biases supported. Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p weights. @@ -127,5 +129,5 @@ private: bool _is_prepared; bool _is_activationlayer_enabled; }; -} +} // namespace arm_compute #endif /* ARM_COMPUTE_NEWINOGRADCONVOLUTIONLAYER_H */ |