diff options
author | Annop Wongwathanarat <annop.wongwathanarat@arm.com> | 2023-03-01 15:19:50 +0000 |
---|---|---|
committer | Annop Wongwathanarat <annop.wongwathanarat@arm.com> | 2023-03-03 12:30:16 +0000 |
commit | adfcacc8e39888a9a62e33c178041642d0a3047a (patch) | |
tree | 3d4998c8c18129cfd38ad7faab22d5002caadbab /arm_compute/runtime/NEON | |
parent | 1fe48cafde21a316011fff32a5b0f98a74fbe2b9 (diff) | |
download | ComputeLibrary-adfcacc8e39888a9a62e33c178041642d0a3047a.tar.gz |
Add weights_info as optional input for NEDeconvolutionLayer
This is so that we can leverage fixed format kernel when
using gemm convolution method.
Partially resolves: [ONCPUML-1129]
Change-Id: I61ffa74f5cd9d75579dbc1f9aa187371f855e932
Signed-off-by: Annop Wongwathanarat <annop.wongwathanarat@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9248
Reviewed-by: Jakub Sujak <jakub.sujak@arm.com>
Reviewed-by: Gunes Bayir <gunes.bayir@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'arm_compute/runtime/NEON')
-rw-r--r-- | arm_compute/runtime/NEON/functions/NEDeconvolutionLayer.h | 19 |
1 files changed, 12 insertions, 7 deletions
diff --git a/arm_compute/runtime/NEON/functions/NEDeconvolutionLayer.h b/arm_compute/runtime/NEON/functions/NEDeconvolutionLayer.h index 869df69f11..cdc3a636b0 100644 --- a/arm_compute/runtime/NEON/functions/NEDeconvolutionLayer.h +++ b/arm_compute/runtime/NEON/functions/NEDeconvolutionLayer.h @@ -110,12 +110,14 @@ public: * @param[in] bias Optional, ignored if NULL. The biases have one dimension. * Data type supported: Data types supported: S32 for QASYMM8/QASYMM8_SIGNED input, F32 for F32 input, F16 for F16 input. * @param[out] output Output tensor. The output has the same number of dimensions as the @p input. - * @param[in] info Contains padding and policies to be used in the deconvolution, this is decribed in @ref PadStrideInfo. + * @param[in] info Contains padding and policies to be used in the deconvolution, this is described in @ref PadStrideInfo. * @param[in] enable_fast_math (Optional) Enable fast math computation. In case this flag were set, the function could dispatch the fastest implementation - * available which may introduce a drop of accuracy as well. Default is false + * available which may introduce a drop of accuracy as well. Default is false + * @param[in] weights_info (Optional) Specifies the weight format. Default is unspecified. This parameter can be used to specify the weight format that is optimal for + * the GEMM convolution. * */ - void configure(ITensor *input, const ITensor *weights, const ITensor *bias, ITensor *output, const PadStrideInfo &info, bool enable_fast_math = false); + void configure(ITensor *input, const ITensor *weights, const ITensor *bias, ITensor *output, const PadStrideInfo &info, bool enable_fast_math = false, const WeightsInfo &weights_info = WeightsInfo()); /** Static function to check if given info will lead to a valid configuration of @ref NEDeconvolutionLayer * * @param[in] input Input tensor info. 3 lower dimensions represent a single input, and an optional 4th dimension for batch of inputs. @@ -124,13 +126,16 @@ public: * Data type supported: Same as @p input, also could be QSYMM8_PER_CHANNEL if input is QASYMM8/QASYMM8_SIGNED. * @param[in] bias (Optional) The biases have one dimension. Data type supported: Data types supported: S32 for QASYMM8/QASYMM8_SIGNED input, F32 for F32 input, F16 for F16 input. * @param[in] output Output tensor info. The output has the same number of dimensions as the @p input. - * @param[in] info Contains padding and policies to be used in the deconvolution, this is decribed in @ref PadStrideInfo. + * @param[in] info Contains padding and policies to be used in the deconvolution, this is described in @ref PadStrideInfo. * @param[in] enable_fast_math (Optional) Enable fast math computation. In case this flag were set, the function could dispatch the fastest implementation - * available which may introduce a drop of accuracy as well. Default is false + * available which may introduce a drop of accuracy as well. Default is false + * @param[in] weights_info (Optional) Specifies the weight format. Default is unspecified. This parameter can be used to specify the weight format that is optimal for + * the GEMM convolution. * * @return a status */ - static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *bias, const ITensorInfo *output, const PadStrideInfo &info, bool enable_fast_math = false); + static Status validate(const ITensorInfo *input, const ITensorInfo *weights, const ITensorInfo *bias, const ITensorInfo *output, const PadStrideInfo &info, + bool enable_fast_math = false, const WeightsInfo &weights_info = WeightsInfo()); // Inherited methods overridden: void run() override; @@ -150,5 +155,5 @@ private: bool _is_prepared; bool _do_upsampling; }; -} // arm_compute +} // namespace arm_compute #endif /* ARM_COMPUTE_NEDECONVOLUTIONLAYER_H */ |