Compute Library
18.05
|
Interface for the kernel to run a 3x3 depthwise convolution on a tensor. More...
#include <NEDepthwiseConvolutionLayer3x3Kernel.h>
Public Member Functions | |
const char * | name () const override |
Name of the kernel. More... | |
NEDepthwiseConvolutionLayer3x3Kernel () | |
Default constructor. More... | |
NEDepthwiseConvolutionLayer3x3Kernel (const NEDepthwiseConvolutionLayer3x3Kernel &)=delete | |
Prevent instances of this class from being copied (As this class contains pointers) More... | |
NEDepthwiseConvolutionLayer3x3Kernel & | operator= (const NEDepthwiseConvolutionLayer3x3Kernel &)=delete |
Prevent instances of this class from being copied (As this class contains pointers) More... | |
NEDepthwiseConvolutionLayer3x3Kernel (NEDepthwiseConvolutionLayer3x3Kernel &&)=default | |
Default Move Constructor. More... | |
NEDepthwiseConvolutionLayer3x3Kernel & | operator= (NEDepthwiseConvolutionLayer3x3Kernel &&)=default |
Default move assignment operator. More... | |
void | configure (const ITensor *input, const ITensor *weights, ITensor *output, const PadStrideInfo &conv_info, unsigned int depth_multiplier=1, DataLayout data_layout=DataLayout::NCHW) |
Initialize the function's source, destination, conv and border_size. More... | |
void | generate_convolver () |
Generates the convolver object. More... | |
void | run (const Window &window, const ThreadInfo &info) override |
Execute the kernel on the passed window. More... | |
BorderSize | border_size () const override |
The size of the border for that kernel. More... | |
Public Member Functions inherited from ICPPKernel | |
virtual | ~ICPPKernel ()=default |
Default destructor. More... | |
Public Member Functions inherited from IKernel | |
IKernel () | |
Constructor. More... | |
virtual | ~IKernel ()=default |
Destructor. More... | |
virtual bool | is_parallelisable () const |
Indicates whether or not the kernel is parallelisable. More... | |
const Window & | window () const |
The maximum window the kernel can be executed on. More... | |
Static Public Member Functions | |
static bool | is_optimized_execution_possible (TensorShape input_shape, PadStrideInfo conv_info, DataType dt, unsigned int depth_multiplier=1, DataLayout data_layout=DataLayout::NCHW) |
Static method that checks if optimized execution is supported for the given parameters. More... | |
Interface for the kernel to run a 3x3 depthwise convolution on a tensor.
Definition at line 37 of file NEDepthwiseConvolutionLayer3x3Kernel.h.
Default constructor.
Referenced by NEDepthwiseConvolutionLayer3x3Kernel::name().
|
delete |
Prevent instances of this class from being copied (As this class contains pointers)
Default Move Constructor.
|
overridevirtual |
The size of the border for that kernel.
Reimplemented from IKernel.
Referenced by NEDepthwiseConvolutionLayer3x3Kernel::name().
void configure | ( | const ITensor * | input, |
const ITensor * | weights, | ||
ITensor * | output, | ||
const PadStrideInfo & | conv_info, | ||
unsigned int | depth_multiplier = 1 , |
||
DataLayout | data_layout = DataLayout::NCHW |
||
) |
Initialize the function's source, destination, conv and border_size.
[in] | input | Source tensor. DataType supported: QASYMM8, F32. |
[in] | weights | Weights tensor. This is a 3D tensor with dimensions [3, 3, IFM]. Data type supported: Same as input . |
[out] | output | Destination tensor. Data type supported: Same as input . |
[in] | conv_info | Padding and stride information to use for the convolution. |
[in] | depth_multiplier | (Optional) Multiplier to apply to the input's depth in order to retrieve the output's depth. Defaults to 1. |
[in] | data_layout | (Optional) Data layout of the input and weights tensor |
Referenced by NEDepthwiseConvolutionLayer3x3Kernel::name().
void generate_convolver | ( | ) |
Generates the convolver object.
Referenced by NEDepthwiseConvolutionLayer3x3Kernel::name().
|
static |
Static method that checks if optimized execution is supported for the given parameters.
[in] | input_shape | Input shape |
[in] | conv_info | Padding and stride information to use for the convolution. |
[in] | dt | Data type of the input and weights |
[in] | data_layout | (Optional) Data layout of the input and weights tensor |
[in] | depth_multiplier | (Optional) Multiplier to apply to the input's depth in order to retrieve the output's depth. Defaults to 1. |
Referenced by arm_compute::test::validation::DATA_TEST_CASE(), and NEDepthwiseConvolutionLayer3x3Kernel::name().
|
inlineoverridevirtual |
Name of the kernel.
Implements ICPPKernel.
Definition at line 40 of file NEDepthwiseConvolutionLayer3x3Kernel.h.
References NEDepthwiseConvolutionLayer3x3Kernel::border_size(), NEDepthwiseConvolutionLayer3x3Kernel::configure(), arm_compute::test::validation::conv_info, NEDepthwiseConvolutionLayer3x3Kernel::generate_convolver(), arm_compute::test::validation::info(), NEDepthwiseConvolutionLayer3x3Kernel::is_optimized_execution_possible(), arm_compute::NCHW, NEDepthwiseConvolutionLayer3x3Kernel::NEDepthwiseConvolutionLayer3x3Kernel(), NEDepthwiseConvolutionLayer3x3Kernel::operator=(), NEDepthwiseConvolutionLayer3x3Kernel::run(), and IKernel::window().
|
delete |
Prevent instances of this class from being copied (As this class contains pointers)
Referenced by NEDepthwiseConvolutionLayer3x3Kernel::name().
|
default |
Default move assignment operator.
|
overridevirtual |
Execute the kernel on the passed window.
[in] | window | Region on which to execute the kernel. (Must be a region of the window returned by window()) |
[in] | info | Info about executing thread and CPU. |
Implements ICPPKernel.
Referenced by NEDepthwiseConvolutionLayer3x3Kernel::name().