diff options
| author | Vidhya Sudhan Loganathan <vidhyasudhan.loganathan@arm.com> | 2018-07-04 09:34:00 +0100 |
|---|---|---|
| committer | Anthony Barbier <anthony.barbier@arm.com> | 2018-11-02 16:54:10 +0000 |
| commit | 7485d5a62685cb745ab50e970adb722cb71557ac (patch) | |
| tree | ba01b99ca466c93edc9a3f8c1e34394ff84be060 /arm_compute/core/NEON/kernels/NEDirectConvolutionLayerOutputStageKernel.h | |
| parent | 014333d73883c3872e458cedda5ccef586a7ccd4 (diff) | |
| download | ComputeLibrary-7485d5a62685cb745ab50e970adb722cb71557ac.tar.gz | |
COMPMID-970 : Remove QS8 / QS16 support
Removed fixed point related code.
Change-Id: I487acf138dace3b0450e0d72ca7071eaec254566
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/137678
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Diffstat (limited to 'arm_compute/core/NEON/kernels/NEDirectConvolutionLayerOutputStageKernel.h')
| -rw-r--r-- | arm_compute/core/NEON/kernels/NEDirectConvolutionLayerOutputStageKernel.h | 8 |
1 files changed, 4 insertions, 4 deletions
diff --git a/arm_compute/core/NEON/kernels/NEDirectConvolutionLayerOutputStageKernel.h b/arm_compute/core/NEON/kernels/NEDirectConvolutionLayerOutputStageKernel.h index 77711d7ecd..7fd1d70374 100644 --- a/arm_compute/core/NEON/kernels/NEDirectConvolutionLayerOutputStageKernel.h +++ b/arm_compute/core/NEON/kernels/NEDirectConvolutionLayerOutputStageKernel.h @@ -55,10 +55,10 @@ public: /** Set the accumulate buffer and the biases of the kernel. * * @param[in, out] input Input to add the bias to. If @p output is not specified then accumulation is done in-place. - * Data type supported: QS16/QS32/F16/F32 + * Data type supported: QS32/F16/F32 * @param[in] bias (Optional) The shared bias tensor to add. It must be 1D Tensor. Data type supported: Same as @p input * @param[out] output (Optional) If the output tensor is specified the accumulation is done out-of-place. (Defaults to nullptr) - * Data type supported: QS8/QS16/F16/F32 + * Data type supported: F16/F32 * @param[in] result_fixedpoint_multiplier (Optional)Fixed point value to be multiplied to each element of the input matrix when once the result_offset has been add * @param[in] result_shift (Optional)Integer value used to round to nearest division by a power-of-two the result after the fixed point multiplication * @param[in] result_offset_after_shift (Optional)Offset to be applied to result before converting it back to QASYMM8 @@ -68,10 +68,10 @@ public: /** Static function to check if given info will lead to a valid configuration of @ref NEDirectConvolutionLayerOutputStageKernel * * @param[in] input Input to add the bias to. If @p output is not specified then accumulation is done in-place. - * Data type supported: QS16/QS32/F16/F32 + * Data type supported: QS32/F16/F32 * @param[in] bias (Optional) The shared bias tensor to add. It must be 1D Tensor. Data type supported: Same as @p input * @param[in] output (Optional) If the output tensor is specified the accumulation is done out-of-place. (Defaults to nullptr) - * Data type supported: QS8/QS16/F16/F32 + * Data type supported: F16/F32 * @return a status */ static Status validate(const ITensorInfo *input, const ITensorInfo *bias = nullptr, const ITensorInfo *output = nullptr); |