Interface to convert the 2D Fully Connected weights from NCHW to NHWC or vice versa. More...

Namespaces
	kernels

Data Structures
class	ClAbs
	Basic function to get the absolute value of an src tensor. More...

class	ClActivation
	Basic function to run kernels::ClActivationKernel. More...

class	ClAdd
	Basic function to run arithmetic addition. More...

class	CLAuxTensorHandler

class	ClCast
	Basic function to run kernels::ClCastKernel. More...

class	ClComplexMul
	Basic function to run opencl::kernels::ClComplexMulKernel. More...

class	ClConcatenate
	Basic function to execute concatenate tensors along a given axis. More...

class	ClConv2d
	Basic function to compute the convolution layer. More...

class	ClConvertFullyConnectedWeights
	Basic function to run kernels::ClConvertFullyConnectedWeightsKernel. More...

class	ClCopy
	Basic function to run kernels::ClCopyKernel. More...

class	ClCrop
	Basic function to run kernels::ClCropKernel. More...

class	ClDequantize
	Basic function to run kernels::ClDequantizeKernel that dequantizes an input tensor. More...

class	ClDirectConv2d
	Basic function to simulate a directly convolution layer. More...

class	ClDirectConv3d
	Basic function to simulate a directly convolution layer with 3 spatial dimensions. More...

class	ClElementwiseDivision
	Basic function to run opencl::kernels::ClArithmeticKernel for division. More...

class	ClElementwiseMax
	Basic function to run opencl::kernels::ClArithmeticKernel for max. More...

class	ClElementwiseMin
	Basic function to run opencl::kernels::ClArithmeticKernel for min. More...

class	ClElementwisePower
	Basic function to run opencl::kernels::ClArithmeticKernel for power. More...

class	ClElementwiseSquaredDiff
	Basic function to run opencl::kernels::ClArithmeticKernel for squared difference. More...

class	ClExp
	Basic function to perform exponential on an src tensor. More...

class	ClFill
	Basic function to run kernels::ClFillKernel. More...

class	ClFlatten
	Basic function to flatten a given input. More...

class	ClFloor
	Basic function to run kernels::ClFloorKernel. More...

class	ClFullyConnected
	Basic function to compute a Fully Connected layer on OpenCL. More...

class	ClGemm
	Basic function to execute GEMM on OpenCL. More...

class	ClGemmConv2d
	Basic function to compute the convolution layer. More...

class	ClGemmLowpMatrixMultiplyCore
	Basic function to execute GEMMLowpMatrixMultiplyCore on OpenCL. More...

class	ClGemmLowpOutputStage
	Basic function to execute GEMMLowpQuantizeDown kernels on CL. More...

class	ClIndirectConv2d
	Basic function to execute indirect convolution on OpenCL. More...

class	ClKernelLibrary
	ClKernelLibrary contains all the OpenCL kernels that are used throughout the library. More...

class	ClLog
	Basic function to perform elementwise log on an src tensor. More...

class	ClLogicalNot
	Basic function to run kernels::ClElementWiseUnaryKernel for NOT operation. More...

class	ClMatMul
	Basic operator to execute BatchMatMul on OpenCL. More...

class	ClMul
	Basic function to run opencl::kernels::ClMulKernel. More...

class	ClNeg
	Basic function to negate an src tensor. More...

class	ClPermute
	Basic function to run kernels::ClPermuteKernel. More...

class	ClPool2d
	Basic function to simulate a pooling layer with the specified pooling operation. More...

class	ClPool3d
	Basic function to simulate a pooling layer with the specified pooling operation. More...

class	ClPRelu
	Basic operator to run arm_compute::opencl::kernels::ClArithmeticKernel for PRELU. More...

class	ClQuantize
	Basic function to run kernels::ClQuantizeKernel that dequantizes an input tensor. More...

class	ClReshape
	Basic function to run kernels::ClReshapeKernel. More...

class	ClRound
	Basic function to get the round (to the nearest even) value of an src tensor. More...

class	ClRsqrt
	Basic function to perform inverse square root on an src tensor. More...

class	ClScale
	Basic function to simulate a scale layer. More...

class	ClSin
	Basic function to calculate sine of an src tensor. More...

class	ClSoftmax

class	ClSub
	Basic function to run arithmetic subtraction. More...

class	ClTranspose
	Basic function to run kernels::ClTransposeKernel. More...

class	ClTransposedConvolution
	Basic function to simulate a directly convolution layer. More...

class	ClWinogradConv2d
	Basic function to execute Winograd-based convolution on OpenCL. More...

Typedefs
using	ClCompileContext = arm_compute::CLCompileContext

using	IClKernel = arm_compute::ICLKernel

using	IClOperator = experimental::ICLOperator

using	KernelType = kernels::ClArithmeticKernel

Detailed Description

Interface to convert the 2D Fully Connected weights from NCHW to NHWC or vice versa.

Note

This function can be applied to the 2D weights used by a Fully Connected layer if:

It follows a Convolution layer
The data layout used by the network does not match the one the model has been trained in.

This function assumes the weights are already reshaped (transposed)

Typedef Documentation

◆ ClCompileContext

using ClCompileContext = arm_compute::CLCompileContext

Definition at line 33 of file ClCompileContext.h.

◆ IClKernel

using IClKernel = arm_compute::ICLKernel

Definition at line 34 of file IClKernel.h.

◆ IClOperator

using IClOperator = experimental::ICLOperator

Definition at line 34 of file IClOperator.h.

◆ KernelType

using KernelType = kernels::ClArithmeticKernel

Definition at line 34 of file ClPRelu.cpp.

Namespaces