diff options
| author | Frank Lei <frank.lei@arm.com> | 2018-02-01 14:47:14 +0800 |
|---|---|---|
| committer | Anthony Barbier <anthony.barbier@arm.com> | 2018-11-02 16:47:40 +0000 |
| commit | 4406fd6cc4abded564d3791324e1f48bdfd34273 (patch) | |
| tree | 22fe402fe9ac7ca338df49e9eccd6eb1587ae875 /arm_compute/runtime/GLES_COMPUTE/functions/GCDirectConvolutionLayer.h | |
| parent | 898d399a0f62c15612a52df4bff5018e783214e4 (diff) | |
| download | ComputeLibrary-4406fd6cc4abded564d3791324e1f48bdfd34273.tar.gz | |
APPBROWSER-391: Fix GLES COMPUTE alignment issues
APPBROWSER-402: Performance optimization for squeezenet/xray model
Change-Id: If31b186b99a6d6087164019fe94d3ac9279e3204
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/119526
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Diffstat (limited to 'arm_compute/runtime/GLES_COMPUTE/functions/GCDirectConvolutionLayer.h')
| -rw-r--r-- | arm_compute/runtime/GLES_COMPUTE/functions/GCDirectConvolutionLayer.h | 19 |
1 files changed, 10 insertions, 9 deletions
diff --git a/arm_compute/runtime/GLES_COMPUTE/functions/GCDirectConvolutionLayer.h b/arm_compute/runtime/GLES_COMPUTE/functions/GCDirectConvolutionLayer.h index c6b948be1f..976aee7b83 100644 --- a/arm_compute/runtime/GLES_COMPUTE/functions/GCDirectConvolutionLayer.h +++ b/arm_compute/runtime/GLES_COMPUTE/functions/GCDirectConvolutionLayer.h @@ -53,16 +53,17 @@ public: GCDirectConvolutionLayer(); /** Set the input and output tensors. * - * @param[in] input Source tensor. 3 lower dimensions represent a single input [width, height, IFM], - * while every optional dimension from 4 and above represent a batch of inputs. - * Data types supported: F16/F32. - * @param[in] weights Weights tensor. Weights are 4D tensor with dimensions [kernel_x, kernel_y, IFM, OFM]. Data type supported:Same as @p input. - * @param[in] biases Biases tensor. Shared biases supported. Biases are 1D tensor with dimensions [OFM]. Data type supported:Same as @p input. - * @param[out] output Destination tensor. 3 lower dimensions represent a single output [width, height, OFM], while the rest represent batch of outputs. - * Data types supported: Same as @p input. - * @param[in] conv_info Contains padding and stride information described in @ref PadStrideInfo. + * @param[in,out] input Source tensor. 3 lower dimensions represent a single input [width, height, IFM], + * while every optional dimension from 4 and above represent a batch of inputs. + * Data types supported: F16/F32. + * input will be written to only if it is currently left aligned. + * @param[in] weights Weights tensor. Weights are 4D tensor with dimensions [kernel_x, kernel_y, IFM, OFM]. Data type supported:Same as @p input. + * @param[in] biases Biases tensor. Shared biases supported. Biases are 1D tensor with dimensions [OFM]. Data type supported:Same as @p input. + * @param[out] output Destination tensor. 3 lower dimensions represent a single output [width, height, OFM], while the rest represent batch of outputs. + * Data types supported: Same as @p input. + * @param[in] conv_info Contains padding and stride information described in @ref PadStrideInfo. */ - void configure(const IGCTensor *input, const IGCTensor *weights, const IGCTensor *biases, IGCTensor *output, const PadStrideInfo &conv_info); + void configure(IGCTensor *input, const IGCTensor *weights, const IGCTensor *biases, IGCTensor *output, const PadStrideInfo &conv_info); // Inherited methods overridden: void run() override final; |