diff options
| author | Michele Di Giorgio <michele.digiorgio@arm.com> | 2017-06-21 15:36:24 +0100 |
|---|---|---|
| committer | Anthony Barbier <anthony.barbier@arm.com> | 2018-09-17 14:15:39 +0100 |
| commit | ab0a77edcb9f48de2aad216323b791d0dd95a3cd (patch) | |
| tree | dab2444a8f4ef70db08c7561b4b5f92648971a59 /arm_compute/core/CL/kernels/CLPixelWiseMultiplicationKernel.h | |
| parent | ccc65d44a53eaa61c718cbc4d826c811e2ccebda (diff) | |
| download | ComputeLibrary-ab0a77edcb9f48de2aad216323b791d0dd95a3cd.tar.gz | |
COMPMID-409: Add support for QS8 and QS16 CLPixelWiseMultiplication.
Change-Id: I7f66d49d746ba9fb6e726ccab83d3a97b8ddef80
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/78491
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Diffstat (limited to 'arm_compute/core/CL/kernels/CLPixelWiseMultiplicationKernel.h')
| -rw-r--r-- | arm_compute/core/CL/kernels/CLPixelWiseMultiplicationKernel.h | 8 |
1 files changed, 4 insertions, 4 deletions
diff --git a/arm_compute/core/CL/kernels/CLPixelWiseMultiplicationKernel.h b/arm_compute/core/CL/kernels/CLPixelWiseMultiplicationKernel.h index 6fbbe95219..a5909d1511 100644 --- a/arm_compute/core/CL/kernels/CLPixelWiseMultiplicationKernel.h +++ b/arm_compute/core/CL/kernels/CLPixelWiseMultiplicationKernel.h @@ -49,11 +49,11 @@ public: CLPixelWiseMultiplicationKernel &operator=(CLPixelWiseMultiplicationKernel &&) = default; /** Initialise the kernel's input, output and border mode. * - * @param[in] input1 An input tensor. Data types supported: U8, S16, F16, F32. - * @param[in] input2 An input tensor. Data types supported: U8, S16, F16, F32. - * @param[out] output The output tensor, Data types supported: U8 (Only if both inputs are U8), S16, F16, F32. + * @param[in] input1 An input tensor. Data types supported: U8/QS8/QS16/S16/F16/F32. + * @param[in] input2 An input tensor. Data types supported: same as @p input1. + * @param[out] output The output tensor, Data types supported: same as @p input1. Note: U8 (QS8, QS16) requires both inputs to be U8 (QS8, QS16). * @param[in] scale Scale to apply after multiplication. - * Scale must be positive and its value must be either 1/255 or 1/2^n where n is between 0 and 15. + * Scale must be positive and its value must be either 1/255 or 1/2^n where n is between 0 and 15. For QS8 and QS16 scale must be 1. * @param[in] overflow_policy Overflow policy. Supported overflow policies: Wrap, Saturate * @param[in] rounding_policy Rounding policy. Supported rounding modes: to zero, to nearest even. */ |