diff options
author | Georgios Pinitas <georgios.pinitas@arm.com> | 2020-07-13 21:21:33 +0100 |
---|---|---|
committer | Georgios Pinitas <georgios.pinitas@arm.com> | 2020-07-14 14:28:46 +0000 |
commit | 4667dddc0ed403c636348294cd7f70261e5540cf (patch) | |
tree | 177b74f377dcbb32cf8a83d407c633df255665a0 /arm_compute/core/NEON/kernels/NEBatchConcatenateLayerKernel.h | |
parent | 2232a201a9f72de483c12a7857c5f08b81cf7396 (diff) | |
download | ComputeLibrary-4667dddc0ed403c636348294cd7f70261e5540cf.tar.gz |
COMPMID-3374: Remove memory state from NEConcatenateLayer kernels
* Allow the following kernels to accept backing memory at run-time:
* NEBatchConcatenateLayerKernel
* NEDepthConcatenateLayerKernel
* NEHeightConcatenateLayerKernel
* NEWidthConcatenateLayerKernel
* Allow the following functions to accept backing memory at run-time:
* NEConcatenateLayer
Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com>
Change-Id: Ib0b6714cff7f06a52dc74d294bc3e0d72a1c2419
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3569
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'arm_compute/core/NEON/kernels/NEBatchConcatenateLayerKernel.h')
-rw-r--r-- | arm_compute/core/NEON/kernels/NEBatchConcatenateLayerKernel.h | 11 |
1 files changed, 5 insertions, 6 deletions
diff --git a/arm_compute/core/NEON/kernels/NEBatchConcatenateLayerKernel.h b/arm_compute/core/NEON/kernels/NEBatchConcatenateLayerKernel.h index 267211fad6..f397a29b48 100644 --- a/arm_compute/core/NEON/kernels/NEBatchConcatenateLayerKernel.h +++ b/arm_compute/core/NEON/kernels/NEBatchConcatenateLayerKernel.h @@ -56,15 +56,15 @@ public: ~NEBatchConcatenateLayerKernel() = default; /** Initialise the kernel's inputs and output * - * @param[in] input Input tensor. Data types supported: All. + * @param[in] input Input tensor info. Data types supported: All. * @param[in] batch_offset The offset on axis # 3. - * @param[in,out] output Output tensor. Data types supported: Same as @p input. + * @param[in,out] output Output tensor info. Data types supported: Same as @p input. * * @note: The output tensor's low two dimensions can't be smaller than the input one's. * @note: The gaps between the two lowest dimensions of input and output need to be divisible by 2. * */ - void configure(const ITensor *input, unsigned int batch_offset, ITensor *output); + void configure(const ITensorInfo *input, unsigned int batch_offset, ITensorInfo *output); /** Static function to check if given info will lead to a valid configuration of @ref NEBatchConcatenateLayerKernel * * @param[in] input Input tensor info. Data types supported: All. @@ -76,15 +76,14 @@ public: static Status validate(const ITensorInfo *input, unsigned int batch_offset, const ITensorInfo *output); // Inherited methods overridden: - void run(const Window &window, const ThreadInfo &info) override; + void run_op(const InputTensorMap &inputs, const OutputTensorMap &outputs, + const Window &window, const ThreadInfo &info) override; private: using BatchConcatFunction = void(const ITensor *in, ITensor *out, unsigned int batch_offset, const Window &window); private: BatchConcatFunction *_func; - const ITensor *_input; - ITensor *_output; unsigned int _batch_offset; }; } // namespace arm_compute |