diff options
author | Sheri Zhang <sheri.zhang@arm.com> | 2020-02-25 15:57:21 +0000 |
---|---|---|
committer | Sheri Zhang <sheri.zhang@arm.com> | 2020-03-09 15:25:17 +0000 |
commit | 0cdbda5e51e6ef9e03017231e56ee85ede69bb9a (patch) | |
tree | e46c17ce06ef0990336834c32ff2488592abc0ec /arm_compute/runtime/CL/functions/CLGEMMDeconvolutionLayer.h | |
parent | f5f2391f0d925f2a8d0833114f63bd8cb1da27b1 (diff) | |
download | ComputeLibrary-0cdbda5e51e6ef9e03017231e56ee85ede69bb9a.tar.gz |
COMPMID-2789: Add support for QASYMM8_SIGNED in CLGEMMDeconvolutionLayer
Change-Id: I7e3bcb01025e827f6f62491749c691c205ee7481
Signed-off-by: Sheri Zhang <sheri.zhang@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/2844
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'arm_compute/runtime/CL/functions/CLGEMMDeconvolutionLayer.h')
-rw-r--r-- | arm_compute/runtime/CL/functions/CLGEMMDeconvolutionLayer.h | 28 |
1 files changed, 15 insertions, 13 deletions
diff --git a/arm_compute/runtime/CL/functions/CLGEMMDeconvolutionLayer.h b/arm_compute/runtime/CL/functions/CLGEMMDeconvolutionLayer.h index 3df9205f48..01687b69ec 100644 --- a/arm_compute/runtime/CL/functions/CLGEMMDeconvolutionLayer.h +++ b/arm_compute/runtime/CL/functions/CLGEMMDeconvolutionLayer.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2019 ARM Limited. + * Copyright (c) 2019-2020 ARM Limited. * * SPDX-License-Identifier: MIT * @@ -68,7 +68,7 @@ class ICLTensor; * This function calls the following OpenCL kernels/functions: * * -# @ref CLGEMMLowpMatrixMultiplyCore - * -# @ref CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint + * -# @ref CLGEMMLowpOutputStage * -# @ref CLPermute * -# @ref CLPermute * -# @ref CLReshapeLayer @@ -91,7 +91,8 @@ public: CLGEMMDeconvolutionLayer &operator=(CLGEMMDeconvolutionLayer &&) = default; /** Set the input, weights, biases and output tensors. * - * @param[in,out] input Input tensor. 3 lower dimensions represent a single input, and an optional 4th dimension for batch of inputs. Data types supported: F16/F32. Data layout supported: NHWC + * @param[in,out] input Input tensor. 3 lower dimensions represent a single input, and an optional 4th dimension for batch of inputs. + * Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32. Data layout supported: NHWC * @param[in] weights The 4d weights with dimensions [width, height, IFM, OFM]. Data type supported: Same as @p input. Data layout supported: same as @p input. * @param[in] bias (Optional) The biases have one dimension. Data type supported: Same as @p input. Data layout supported: same as @p input. * @param[out] output Output tensor. The output has the same number of dimensions as the @p input. Data layout supported: same as @p input. @@ -100,7 +101,8 @@ public: void configure(const ICLTensor *input, const ICLTensor *weights, const ICLTensor *bias, ICLTensor *output, const PadStrideInfo &deconv_info); /** Static function to check if given info will lead to a valid configuration of @ref CLDeconvolutionLayer * - * @param[in] input Input tensor info. 3 lower dimensions represent a single input, and an optional 4th dimension for batch of inputs. Data types supported: F16/F32. Data layout supported: NHWC + * @param[in] input Input tensor info. 3 lower dimensions represent a single input, and an optional 4th dimension for batch of inputs. + * Data types supported: QASYMM8/QASYMM8_SIGNED/F16/F32. Data layout supported: NHWC * @param[in] weights The 4d weights info with dimensions [width, height, IFM, OFM]. Data type supported: Same as @p input. Data layout supported: same as @p input. * @param[in] bias (Optional) The biases have one dimension. Data type supported: Same as @p input. Data layout supported: same as @p input. * @param[in] output Output tensor info. The output has the same number of dimensions as the @p input. Data layout supported: same as @p input. @@ -117,15 +119,15 @@ public: private: MemoryGroup _memory_group; - CLGEMM _mm_gemm; - CLGEMMLowpMatrixMultiplyCore _mm_gemmlowp; - CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint _gemmlowp_output_stage; - CLPermute _permute_input_to_nhwc; - CLPermute _permute_weights_to_nhwc; - CLReshapeLayer _reshape_weights; - CLTranspose _transpose_weights; - CLDeconvolutionReshapeOutputKernel _deconv_reshape; - CLSlice _slice_gemm; + CLGEMM _mm_gemm; + CLGEMMLowpMatrixMultiplyCore _mm_gemmlowp; + CLGEMMLowpOutputStage _gemmlowp_output_stage; + CLPermute _permute_input_to_nhwc; + CLPermute _permute_weights_to_nhwc; + CLReshapeLayer _reshape_weights; + CLTranspose _transpose_weights; + CLDeconvolutionReshapeOutputKernel _deconv_reshape; + CLSlice _slice_gemm; CLTensor _gemmlowp_final; CLTensor _reshaped_weights; |