diff options
| author | Giorgio Arena <giorgio.arena@arm.com> | 2018-06-08 12:24:14 +0100 |
|---|---|---|
| committer | Anthony Barbier <anthony.barbier@arm.com> | 2018-11-02 16:53:09 +0000 |
| commit | 7210fe8bb3fefc13b14ded356a9acb0a57aef277 (patch) | |
| tree | 1d71a7433c2564182df33a5a7c77260c59a282ed /arm_compute/core/CL/kernels/CLWinogradOutputTransformKernel.h | |
| parent | fa23f1102f3e2d41838b8a9b53ab74c24cea5b50 (diff) | |
| download | ComputeLibrary-7210fe8bb3fefc13b14ded356a9acb0a57aef277.tar.gz | |
COMPMID-1204 Add NHWC data format support to Winograd output transform 4x4_5x5
Change-Id: Id6dece059b521e50ef546c3ee2883acedf8e3b1c
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/134760
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'arm_compute/core/CL/kernels/CLWinogradOutputTransformKernel.h')
| -rw-r--r-- | arm_compute/core/CL/kernels/CLWinogradOutputTransformKernel.h | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/arm_compute/core/CL/kernels/CLWinogradOutputTransformKernel.h b/arm_compute/core/CL/kernels/CLWinogradOutputTransformKernel.h index 03e3bf5740..cd46e9813e 100644 --- a/arm_compute/core/CL/kernels/CLWinogradOutputTransformKernel.h +++ b/arm_compute/core/CL/kernels/CLWinogradOutputTransformKernel.h @@ -51,7 +51,7 @@ public: * @note Winograd output transform supports the following configurations: * F(output tile, kernel size):F(2x2, 3x3), F(4x4, 3x3), F(4x4, 5x5) * Strides: only unit strides - * Data Layout: NCHW for all configurations, NHWC for F(4x4, 3x3) + * Data Layout: NCHW for all configurations, NHWC for F(4x4, 3x3) and F(4x4, 5x5) * * @param[in] input Source tensor with shape [C, N, K, batches]. Data types supported: F32. * @param[in] bias Biases tensor. Shared biases supported. Biases are 1D tensor with dimensions [OFM]. It can be a nullptr. Data type supported: as @p input @@ -64,7 +64,7 @@ public: * @note Winograd output transform supports the following configurations: * F(output tile, kernel size):F(2x2, 3x3), F(4x4, 3x3), F(4x4, 5x5) * Strides: only unit strides - * Data Layout: NCHW for all configurations, NHWC for F(4x4, 3x3) + * Data Layout: NCHW for all configurations, NHWC for F(4x4, 3x3) and F(4x4, 5x5) * * @param[in] input Source tensor with shape [C, N, K, batches]. Data types supported: F32. * @param[in] bias Biases tensor. Shared biases supported. Biases are 1D tensor with dimensions [OFM]. It can be a nullptr. Data type supported: as @p input |