diff options
| author | Georgios Pinitas <georgios.pinitas@arm.com> | 2017-07-05 17:02:25 +0100 |
|---|---|---|
| committer | Anthony Barbier <anthony.barbier@arm.com> | 2018-09-17 14:15:39 +0100 |
| commit | ac4e873dad6aa6291fc36aff62047a896db04f6a (patch) | |
| tree | 2c5397c6c04b97cedb88ff809f5d40cbe7dc79c9 /arm_compute/runtime | |
| parent | df24618b53cffed1c574e11e9fd4ba7740f8c009 (diff) | |
| download | ComputeLibrary-ac4e873dad6aa6291fc36aff62047a896db04f6a.tar.gz | |
COMPMID-417: Port DepthConcatenate to QS8/QS16 for NEON/CL.
Change-Id: I3dddae63043c7aa18d908a4fc8abacf3c64f98ca
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/80081
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Steven Niu <steven.niu@arm.com>
Diffstat (limited to 'arm_compute/runtime')
| -rw-r--r-- | arm_compute/runtime/CL/functions/CLDepthConcatenate.h | 9 | ||||
| -rw-r--r-- | arm_compute/runtime/NEON/functions/NEDepthConcatenate.h | 9 |
2 files changed, 10 insertions, 8 deletions
diff --git a/arm_compute/runtime/CL/functions/CLDepthConcatenate.h b/arm_compute/runtime/CL/functions/CLDepthConcatenate.h index 3199936b82..77997f6bd1 100644 --- a/arm_compute/runtime/CL/functions/CLDepthConcatenate.h +++ b/arm_compute/runtime/CL/functions/CLDepthConcatenate.h @@ -29,14 +29,15 @@ #include "arm_compute/core/Window.h" #include "arm_compute/runtime/IFunction.h" +#include "arm_compute/core/CL/kernels/CLDepthConcatenateKernel.h" +#include "arm_compute/core/CL/kernels/CLFillBorderKernel.h" + #include <memory> #include <vector> namespace arm_compute { class ICLTensor; -class CLDepthConcatenateKernel; -class CLFillBorderKernel; /** Basic function to execute concatenate tensors along z axis. This function calls the following kernels: * @@ -51,8 +52,8 @@ public: CLDepthConcatenate(); /** Initialise the kernel's inputs vector and output. * - * @param[in,out] inputs_vector The vectors containing all the tensors to concatenate. Data types supported: F32. - * @param[out] output Output tensor. Data types supported: F32. + * @param[in,out] inputs_vector The vectors containing all the tensors to concatenate. Data types supported: QS8/QS16/F16/F32. + * @param[out] output Output tensor. Data types supported: Same as @p input. */ void configure(std::vector<ICLTensor *> inputs_vector, ICLTensor *output); diff --git a/arm_compute/runtime/NEON/functions/NEDepthConcatenate.h b/arm_compute/runtime/NEON/functions/NEDepthConcatenate.h index 02ff1227c7..cc65099575 100644 --- a/arm_compute/runtime/NEON/functions/NEDepthConcatenate.h +++ b/arm_compute/runtime/NEON/functions/NEDepthConcatenate.h @@ -26,14 +26,15 @@ #include "arm_compute/runtime/IFunction.h" +#include "arm_compute/core/NEON/kernels/NEDepthConcatenateKernel.h" +#include "arm_compute/core/NEON/kernels/NEFillBorderKernel.h" + #include <memory> #include <vector> namespace arm_compute { class ITensor; -class NEDepthConcatenateKernel; -class NEFillBorderKernel; /** Basic function to execute concatenate tensors along z axis. This function calls the following kernels: * @@ -48,8 +49,8 @@ public: NEDepthConcatenate(); /** Initialise the kernel's inputs vector and output. * - * @param[in,out] inputs_vector The vectors containing all the tensors to concatenate. Data types supported: F32. - * @param[out] output Output tensor. Data types supported: F32. + * @param[in,out] inputs_vector The vectors containing all the tensors to concatenate. Data types supported: QS8/QS16/F16/F32. + * @param[out] output Output tensor. Data types supported: Same as @p inputs_vector. */ void configure(std::vector<ITensor *> inputs_vector, ITensor *output); |