diff options
| author | Sang-Hoon Park <sang-hoon.park@arm.com> | 2020-07-07 09:36:09 +0100 |
|---|---|---|
| committer | Sang-Hoon Park <sang-hoon.park@arm.com> | 2020-07-08 10:00:17 +0000 |
| commit | a0205b987509d239b1635024fe8f334a4534f56e (patch) | |
| tree | b1751b74d14073d2c711de486d3f0efb1a4d2c6c /arm_compute/core | |
| parent | f9b595adbdc3f6f51ffa2c1f2aa70d0262d0db2d (diff) | |
| download | ComputeLibrary-a0205b987509d239b1635024fe8f334a4534f56e.tar.gz | |
COMPMID-3574: add logarithm to LogSoftmaxLayer
Missed logarithm for the summation is added to NEON, CL and reference
backends. To avoid complex changes, log softmax layer on CL backend
doesn't support quantized data types. Tests and doxygen comments
are modified accordingly.
Change-Id: Iafd29291be8b81345cb4999b2668dbc3ae0c3345
Signed-off-by: Sang-Hoon Park <sang-hoon.park@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3517
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
Reviewed-by: SiCong Li <sicong.li@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'arm_compute/core')
| -rw-r--r-- | arm_compute/core/CL/kernels/CLSoftmaxLayerKernel.h | 6 |
1 files changed, 3 insertions, 3 deletions
diff --git a/arm_compute/core/CL/kernels/CLSoftmaxLayerKernel.h b/arm_compute/core/CL/kernels/CLSoftmaxLayerKernel.h index b174f493b5..bd544e1537 100644 --- a/arm_compute/core/CL/kernels/CLSoftmaxLayerKernel.h +++ b/arm_compute/core/CL/kernels/CLSoftmaxLayerKernel.h @@ -203,7 +203,7 @@ public: CLLogits1DNormKernel &operator=(CLLogits1DNormKernel &&) = default; /** Set the input and output tensors. * - * @param[in] input Source tensor. Data types supported: S32/F16/F32 + * @param[in] input Source tensor. Data types supported: S32/F16/F32. If this kernel is used for log softmax, only F32/F16 is supported. * @param[in] sum Sum tensor. Dimensions should be dim(input)-1. Data types supported: same as @p input * @param[out] output Destination tensor. Data types supported: QASYMM8 for S32 @p input, or same as @p input * @param[in] info Contains information consumed by kernels for softmax described in @ref SoftmaxKernelInfo. @@ -212,7 +212,7 @@ public: /** Set the input and output tensors. * * @param[in] compile_context The compile context to be used. - * @param[in] input Source tensor. Data types supported: S32/F16/F32 + * @param[in] input Source tensor. Data types supported: S32/F16/F32. If this kernel is used for log softmax, only F32/F16 is supported. * @param[in] sum Sum tensor. Dimensions should be dim(input)-1. Data types supported: same as @p input * @param[out] output Destination tensor. Data types supported: QASYMM8 for S32 @p input, or same as @p input * @param[in] info Contains information consumed by kernels for softmax described in @ref SoftmaxKernelInfo. @@ -220,7 +220,7 @@ public: void configure(const CLCompileContext &compile_context, const ICLTensor *input, const ICLTensor *sum, ICLTensor *output, const SoftmaxKernelInfo &info); /** Static function to check if given info will lead to a valid configuration of @ref CLLogits1DNormKernel * - * @param[in] input Source tensor. Data types supported: S32/F16/F32 + * @param[in] input Source tensor. Data types supported: S32/F16/F32. If this kernel is used for log softmax, only F32/F16 is supported. * @param[in] sum Sum tensor. Dimensions should be dim(input)-1. Data types supported: same as @p input * @param[in] output Destination tensor. Data types supported: QASYMM8 for S32 @p input, or same as @p input * @param[in] info Contains information consumed by kernels for softmax described in @ref SoftmaxKernelInfo. |