Compute Library
 19.08
arm_compute Namespace Reference

Copyright (c) 2017-2018 ARM Limited. More...

Namespaces

 cl_gemm
 
 cl_tuner
 
 detail
 
 gles
 
 graph
 
 graph_utils
 
 helpers
 
 io
 
 logging
 
 misc
 
 quantization
 
 support
 
 test
 
 traits
 
 tuners
 
 utility
 
 utils
 
 wrapper
 

Data Structures

class  AccessWindowAutoPadding
 Dummy access window. More...
 
class  AccessWindowHorizontal
 Implementation of a row access pattern. More...
 
class  AccessWindowRectangle
 Implementation of a rectangular access pattern. More...
 
class  AccessWindowStatic
 Implementation of a static rectangular access pattern. More...
 
class  AccessWindowTranspose
 Implementation of a XY-transpose access pattern. More...
 
class  AccessWindowVertical
 Implementation of a column access pattern. More...
 
class  ActivationLayerInfo
 Activation Layer Information class. More...
 
class  Allocator
 Default malloc allocator implementation. More...
 
class  Array
 Basic implementation of the IArray interface which allocates a static number of T values. More...
 
struct  BlobInfo
 Meta-data information for each blob. More...
 
class  BlobLifetimeManager
 Concrete class that tracks the lifetime of registered tensors and calculates the systems memory requirements in terms of blobs. More...
 
class  BlobMemoryPool
 Blob memory pool. More...
 
struct  BorderSize
 Container for 2D border size. More...
 
class  BoundingBoxTransformInfo
 Bounding Box Transform information class. More...
 
class  BoxNMSLimitInfo
 BoxWithNonMaximaSuppressionLimit Information class. More...
 
class  CLAbsLayer
 Basic function to get the absolute value of an input tensor. More...
 
class  CLAbsoluteDifference
 Basic function to run CLAbsoluteDifferenceKernel. More...
 
class  CLAbsoluteDifferenceKernel
 Interface for the absolute difference kernel. More...
 
class  CLAccumulate
 Basic function to run CLAccumulateKernel. More...
 
class  CLAccumulateKernel
 Interface for the accumulate kernel. More...
 
class  CLAccumulateSquared
 Basic function to run CLAccumulateSquaredKernel. More...
 
class  CLAccumulateSquaredKernel
 Interface for the accumulate squared kernel. More...
 
class  CLAccumulateWeighted
 Basic function to run CLAccumulateWeightedKernel. More...
 
class  CLAccumulateWeightedKernel
 Interface for the accumulate weighted kernel. More...
 
class  CLActivationLayer
 Basic function to run CLActivationLayerKernel. More...
 
class  CLActivationLayerKernel
 Interface for the activation layer kernel. More...
 
class  CLArgMinMaxLayer
 Function to calculate the index of the minimum or maximum values in a tensor based on an axis. More...
 
class  CLArithmeticAddition
 Basic function to run CLSaturatedArithmeticOperationKernel for addition. More...
 
class  CLArithmeticDivision
 Basic function to run CLSaturatedArithmeticOperationKernel for division. More...
 
class  CLArithmeticOperationKernel
 
class  CLArithmeticSubtraction
 Basic function to run CLSaturatedArithmeticOperationKernel for subtraction. More...
 
class  CLArray
 CLArray implementation. More...
 
class  CLBatchConcatenateLayerKernel
 Interface for the batch concatenate kernel. More...
 
class  CLBatchNormalizationLayer
 Basic function to run CLNormalizationLayerKernel and simulate a batch normalization layer. More...
 
class  CLBatchNormalizationLayerKernel
 Interface for the BatchNormalization layer kernel. More...
 
class  CLBatchToSpaceLayer
 Basic function to run CLBatchToSpaceLayerKernel. More...
 
class  CLBatchToSpaceLayerKernel
 Interface for the batch to space kernel. More...
 
class  CLBitwiseAnd
 Basic function to run CLBitwiseAndKernel. More...
 
class  CLBitwiseAndKernel
 Interface for the bitwise AND operation kernel. More...
 
class  CLBitwiseNot
 Basic function to run CLBitwiseNotKernel. More...
 
class  CLBitwiseNotKernel
 Interface for the bitwise NOT operation kernel. More...
 
class  CLBitwiseOr
 Basic function to run CLBitwiseOrKernel. More...
 
class  CLBitwiseOrKernel
 Interface for the bitwise OR operation kernel. More...
 
class  CLBitwiseXor
 Basic function to run CLBitwiseXorKernel. More...
 
class  CLBitwiseXorKernel
 Interface for the bitwise XOR operation kernel. More...
 
class  CLBoundingBoxTransform
 Basic function to run CLBoundingBoxTransformKernel. More...
 
class  CLBoundingBoxTransformKernel
 Interface for the bounding box kernel. More...
 
class  CLBox3x3
 Basic function to execute box filter 3x3. More...
 
class  CLBox3x3Kernel
 Interface for the box 3x3 filter kernel. More...
 
class  CLBufferAllocator
 Default OpenCL cl buffer allocator implementation. More...
 
class  CLBufferMemoryRegion
 OpenCL buffer memory region implementation. More...
 
class  CLBuildOptions
 Build options. More...
 
class  CLCannyEdge
 Basic function to execute canny edge on OpenCL. More...
 
class  CLCast
 Basic function to run CLDepthConvertLayerKernel. More...
 
class  CLChannelCombine
 Basic function to run CLChannelCombineKernel to perform channel combination. More...
 
class  CLChannelCombineKernel
 Interface for the channel combine kernel. More...
 
class  CLChannelExtract
 Basic function to run CLChannelExtractKernel to perform channel extraction. More...
 
class  CLChannelExtractKernel
 Interface for the channel extract kernel. More...
 
class  CLChannelShuffleLayer
 Basic function to run CLChannelShuffleLayerKernel. More...
 
class  CLChannelShuffleLayerKernel
 Interface for the channel shuffle kernel. More...
 
class  CLCoarseSVMMemoryRegion
 OpenCL coarse-grain SVM memory region implementation. More...
 
struct  CLCoefficientTable
 Structure for storing Spatial Gradient Matrix and the minimum eigenvalue for each keypoint. More...
 
class  CLCol2ImKernel
 Interface for the col2im reshaping kernel. More...
 
class  CLColorConvert
 Basic function to run CLColorConvertKernel. More...
 
class  CLColorConvertKernel
 Interface for the color convert kernel. More...
 
class  CLComparison
 Basic function to run CLComparisonKernel. More...
 
class  CLComparisonKernel
 Interface for the comparison kernel. More...
 
class  CLComparisonStatic
 Basic function to run CLComparisonKernel. More...
 
class  CLComplexPixelWiseMultiplication
 Basic function to run CLComplexPixelWiseMultiplicationKernel. More...
 
class  CLComplexPixelWiseMultiplicationKernel
 Interface for the complex pixelwise multiplication kernel. More...
 
class  CLComputeAllAnchors
 Basic function to run CLComputeAllAnchorsKernel. More...
 
class  CLComputeAllAnchorsKernel
 Interface for Compute All Anchors kernel. More...
 
class  CLConcatenateLayer
 Basic function to execute concatenate tensors along a given axis. More...
 
class  CLConvertFullyConnectedWeights
 Basic function to run CLConvertFullyConnectedWeightsKernel. More...
 
class  CLConvertFullyConnectedWeightsKernel
 Interface to convert the 2D Fully Connected weights from NCHW to NHWC or vice versa. More...
 
class  CLConvolution3x3
 Basic function to execute convolution of size 3x3. More...
 
class  CLConvolutionKernel
 Interface for the kernel to run an arbitrary size convolution on a tensor. More...
 
class  CLConvolutionLayer
 Basic function to compute the convolution layer. More...
 
class  CLConvolutionLayerReshapeWeights
 Function to reshape and transpose the weights. More...
 
class  CLConvolutionRectangle
 Basic function to execute non-square convolution. More...
 
class  CLConvolutionRectangleKernel
 Kernel for the running convolution on a rectangle matrix. More...
 
class  CLConvolutionSquare
 Basic function to execute square convolution.Currently it supports 5x5, 7x7, 9x9. More...
 
class  CLCopy
 
class  CLCopyKernel
 OpenCL kernel to perform a copy between two tensors. More...
 
class  CLCopyToArrayKernel
 CL kernel to copy keypoints information to ICLKeyPointArray and counts the number of key points. More...
 
class  CLCropKernel
 OpenCL kernel to perform a copy between two tensors. More...
 
class  CLCropResize
 Function to perform cropping and resizing. More...
 
class  CLDeconvolutionLayer
 Basic function to compute the deconvolution layer. More...
 
class  CLDeconvolutionLayerUpsample
 Basic function to execute deconvolution upsample on OpenCL. More...
 
class  CLDeconvolutionLayerUpsampleKernel
 Interface for the Deconvolution layer kernel on OpenCL. More...
 
class  CLDeconvolutionReshapeOutputKernel
 Interface for the OpenCL kernel to be used for reshaping the tensor before returning the result of deconvolution. More...
 
class  CLDepthConcatenateLayerKernel
 Interface for the depth concatenate kernel. More...
 
class  CLDepthConvertLayer
 Basic function to run CLDepthConvertLayerKernel. More...
 
class  CLDepthConvertLayerKernel
 Interface for the depth conversion kernel. More...
 
class  CLDepthToSpaceLayer
 Basic function to run CLDepthToSpaceLayerKernel. More...
 
class  CLDepthToSpaceLayerKernel
 Interface for the depth to space kernel. More...
 
class  CLDepthwiseConvolutionLayer
 Basic function to execute a generic depthwise convolution. More...
 
class  CLDepthwiseConvolutionLayer3x3
 Basic function to execute a depthwise convolution for kernel size 3x3xC (when data layout NCHW) or Cx3x3 (when data layout NHWC). More...
 
class  CLDepthwiseConvolutionLayer3x3NCHWKernel
 Interface for the kernel to run a 3x3 depthwise convolution on a tensor when the data layout is NCHW. More...
 
class  CLDepthwiseConvolutionLayer3x3NHWCKernel
 Interface for the kernel to run a 3x3 depthwise convolution on a tensor when the data layout is NHWC. More...
 
class  CLDepthwiseConvolutionLayerReshapeWeightsGenericKernel
 Interface for the depthwise weights reshape kernel. More...
 
class  CLDepthwiseConvolutionLayerReshapeWeightsKernel
 Interface for the kernel to reshape the weights of depthwise convolution. More...
 
class  CLDepthwiseIm2ColKernel
 Interface for the depthwise im2col reshape kernel. More...
 
class  CLDepthwiseSeparableConvolutionLayer
 Basic function to execute depthwise convolution. More...
 
class  CLDepthwiseVectorToTensorKernel
 Interface for the depthwise vector to tensor kernel. More...
 
class  CLDequantizationLayer
 Basic function to run CLDequantizationLayerKernel that dequantizes an input tensor. More...
 
class  CLDequantizationLayerKernel
 Interface for the dequantization layer kernel. More...
 
class  CLDerivative
 Basic function to execute first order derivative operator. More...
 
class  CLDerivativeKernel
 Interface for the derivative kernel. More...
 
struct  CLDeviceOptions
 OpenCL device options. More...
 
class  CLDilate
 Basic function to execute dilate. More...
 
class  CLDilateKernel
 Interface for the dilate kernel. More...
 
class  CLDirectConvolutionLayer
 Basic function to execute direct convolution function: More...
 
class  CLDirectConvolutionLayerKernel
 Interface for the direct convolution kernel. More...
 
class  CLDirectConvolutionLayerOutputStageKernel
 OpenCL kernel to accumulate the biases, if provided, or downscale in case of quantized input. More...
 
class  CLDirectDeconvolutionLayer
 Function to run the deconvolution layer. More...
 
class  CLDistribution1D
 CLDistribution1D object class. More...
 
class  CLEdgeNonMaxSuppressionKernel
 OpenCL kernel to perform Non-Maxima suppression for Canny Edge. More...
 
class  CLEdgeTraceKernel
 OpenCL kernel to perform Edge tracing. More...
 
class  CLElementwiseMax
 Basic function to run CLArithmeticOperationKernel for max. More...
 
class  CLElementwiseMin
 Basic function to run CLArithmeticOperationKernel for min. More...
 
class  CLElementwiseOperationKernel
 Interface for an element-wise operation kernel. More...
 
class  CLElementwisePower
 Basic function to run CLArithmeticOperationKernel for power. More...
 
class  CLElementwiseSquaredDiff
 Basic function to run CLArithmeticOperationKernel for squared difference. More...
 
class  CLElementWiseUnaryLayerKernel
 Interface for the elementwise unary operator. More...
 
class  CLEqualizeHistogram
 Basic function to execute histogram equalization. More...
 
class  CLErode
 Basic function to execute erode. More...
 
class  CLErodeKernel
 Interface for the erode kernel. More...
 
class  CLExpLayer
 Basic function to perform exponential on an input tensor. More...
 
class  CLFastCorners
 Basic function to execute fast corners. More...
 
class  CLFastCornersKernel
 CL kernel to perform fast corners. More...
 
class  CLFFT1D
 Basic function to execute one dimensional FFT. More...
 
class  CLFFT2D
 Basic function to execute two dimensional FFT. More...
 
class  CLFFTConvolutionLayer
 Basic function to execute FFT-based convolution on OpenCL. More...
 
class  CLFFTDigitReverseKernel
 Interface for the digit reverse operation kernel. More...
 
class  CLFFTRadixStageKernel
 Interface for the FFT radix stage kernel. More...
 
class  CLFFTScaleKernel
 Interface for the inverse fft scale kernel. More...
 
class  CLFillBorder
 Basic function to run CLFillBorderKernel. More...
 
class  CLFillBorderKernel
 Interface for filling the border of a kernel. More...
 
class  CLFineSVMMemoryRegion
 OpenCL fine-grain SVM memory region implementation. More...
 
class  CLFlattenLayer
 Basic function to execute flatten. More...
 
class  CLFlattenLayerKernel
 OpenCL interface for the flatten kernel. More...
 
class  CLFloor
 Basic function to run CLFloorKernel. More...
 
class  CLFloorKernel
 OpenCL kernel to perform a floor operation. More...
 
class  CLFullyConnectedLayer
 Basic function to compute a Fully Connected layer on OpenCL. More...
 
class  CLFullyConnectedLayerReshapeWeights
 Basic function to reshape the weights of Fully Connected layer with OpenCL. More...
 
class  CLFuseBatchNormalization
 Basic function to fuse the batch normalization node to a preceding convolution node. More...
 
class  CLFuseBatchNormalizationKernel
 OpenCL kernel to fuse the batch normalization node to a preceding convolution node. More...
 
class  CLGather
 Basic function to run CLGatherKernel. More...
 
class  CLGatherKernel
 Interface for the kernel to perform tensor reshaping. More...
 
class  CLGaussian3x3
 Basic function to execute gaussian filter 3x3. More...
 
class  CLGaussian3x3Kernel
 Interface for the Gaussian 3x3 filter kernel. More...
 
class  CLGaussian5x5
 Basic function to execute gaussian filter 5x5. More...
 
class  CLGaussian5x5HorKernel
 Interface for the kernel to run the horizontal pass of 5x5 Gaussian filter on a tensor. More...
 
class  CLGaussian5x5VertKernel
 Interface for the kernel to run the vertical pass of 5x5 Gaussian filter on a tensor. More...
 
class  CLGaussianPyramid
 Common interface for all Gaussian pyramid functions. More...
 
class  CLGaussianPyramidHalf
 Basic function to execute gaussian pyramid with HALF scale factor. More...
 
class  CLGaussianPyramidHorKernel
 OpenCL kernel to perform a Gaussian filter and half scaling across width (horizontal pass) More...
 
class  CLGaussianPyramidOrb
 Basic function to execute gaussian pyramid with ORB scale factor. More...
 
class  CLGaussianPyramidVertKernel
 OpenCL kernel to perform a Gaussian filter and half scaling across height (vertical pass) More...
 
class  CLGEMM
 Basic function to execute GEMM on OpenCL. More...
 
class  CLGEMMConvolutionLayer
 Basic function to compute the convolution layer. More...
 
class  CLGEMMDeconvolutionLayer
 Function to run the deconvolution layer through a call to GEMM. More...
 
class  CLGEMMLowpMatrixAReductionKernel
 OpenCL kernel used to compute the row-vectors of sums of all the entries in each row of Matrix A. More...
 
class  CLGEMMLowpMatrixBReductionKernel
 OpenCL kernel used to compute the row-vectors of sums of all the entries in each column of Matrix B. More...
 
class  CLGEMMLowpMatrixMultiplyCore
 Basic function to execute GEMMLowpMatrixMultiplyCore on OpenCL. More...
 
class  CLGEMMLowpMatrixMultiplyKernel
 OpenCL kernel to multiply matrices. More...
 
class  CLGEMMLowpMatrixMultiplyNativeKernel
 OpenCL kernel to multiply matrices with QASYMM8 data type. More...
 
class  CLGEMMLowpMatrixMultiplyReshapedKernel
 OpenCL kernel to multiply matrices when both the input matrices LHS (input0) and RHS (input1) have been reshaped. More...
 
class  CLGEMMLowpMatrixMultiplyReshapedOnlyRHSKernel
 OpenCL kernel to multiply matrices with QASYMM8 data type when only the input matrix RHS (input1) has been reshaped. More...
 
class  CLGEMMLowpOffsetContributionKernel
 OpenCL kernel used to add the offset contribution after CLGEMMLowpMatrixMultiplyKernel. More...
 
class  CLGEMMLowpOffsetContributionOutputStageKernel
 OpenCL kernel used to add the offset contribution after CLGEMMLowpMatrixMultiplyKernel and perform the output stage. More...
 
class  CLGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPoint
 Basic function to execute CLGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPoint on OpenCL. More...
 
class  CLGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel
 CL kernel used to quantize down the int32 accumulator values of GEMMLowp to QSYMM16. More...
 
class  CLGEMMLowpQuantizeDownInt32ToUint8Scale
 Basic function to execute CLGEMMLowpQuantizeDownInt32ToUint8Scale on OpenCL. More...
 
class  CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint
 Basic function to execute CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint on OpenCL. More...
 
class  CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel
 OpenCL kernel used to quantize down the int32 accumulator values of GEMMLowp to QASYMM8. More...
 
class  CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFloat
 Basic function to execute CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFloat on OpenCL. More...
 
class  CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFloatKernel
 OpenCL kernel used to quantize down the int32 accumulator values of GEMMLowp to QASYMM8. More...
 
class  CLGEMMLowpQuantizeDownInt32ToUint8ScaleKernel
 OpenCL kernel used to quantize down the int32 accumulator values of GEMMLowp to QASYMM8. More...
 
class  CLGEMMMatrixAccumulateBiasesKernel
 Interface to add a bias to each row of the input tensor. More...
 
class  CLGEMMMatrixAdditionKernel
 OpenCL kernel to perform the in-place matrix addition between 2 matrices, taking into account that the second matrix might be weighted by a scalar value beta. More...
 
class  CLGEMMMatrixMultiplyKernel
 OpenCL kernel to multiply two input matrices "A" and "B" and add a martix "C" if provided. More...
 
class  CLGEMMMatrixMultiplyNativeKernel
 OpenCL kernel to multiply matrices when neither of the input matrices have been reshaped. More...
 
class  CLGEMMMatrixMultiplyReshapedKernel
 OpenCL kernel to multiply matrices when both the input matrices LHS (input0) and RHS (input1) have been reshaped. More...
 
class  CLGEMMMatrixMultiplyReshapedOnlyRHSKernel
 OpenCL kernel to multiply matrices when only the input matrix RHS (input1) has been reshaped. More...
 
class  CLGEMMMatrixVectorMultiplyKernel
 Interface for the GEMM matrix vector multiply kernel. More...
 
class  CLGEMMReshapeLHSMatrixKernel
 OpenCL kernel to reshape the LHS matrix when performing the matrix multiplication. More...
 
class  CLGEMMReshapeRHSMatrixKernel
 OpenCL kernel to reshape the RHS matrix when performing the matrix multiplication In particular, this kernel splits the input matrix in blocks of size K0xN0 and stores each one in the output matrix unrolling the values. More...
 
class  CLGenerateProposalsLayer
 Basic function to generate proposals for a RPN (Region Proposal Network) More...
 
class  CLGradientKernel
 OpenCL kernel to perform Gradient computation. More...
 
class  CLHarrisCorners
 Basic function to execute harris corners detection. More...
 
class  CLHarrisScoreKernel
 Interface for the harris score kernel. More...
 
class  CLHeightConcatenateLayerKernel
 Interface for the height concatenate kernel. More...
 
class  CLHistogram
 Basic function to execute histogram. More...
 
class  CLHistogramBorderKernel
 Interface to run the histogram kernel to handle the leftover part of image. More...
 
class  CLHistogramKernel
 Interface to run the histogram kernel. More...
 
class  CLHOG
 OpenCL implementation of HOG data-object. More...
 
class  CLHOGBlockNormalizationKernel
 OpenCL kernel to perform HOG block normalization. More...
 
class  CLHOGDescriptor
 Basic function to calculate HOG descriptor. More...
 
class  CLHOGDetector
 Basic function to execute HOG detector based on linear SVM. More...
 
class  CLHOGDetectorKernel
 OpenCL kernel to perform HOG detector kernel using linear SVM. More...
 
class  CLHOGGradient
 Basic function to calculate the gradient for HOG. More...
 
class  CLHOGMultiDetection
 Basic function to detect multiple objects (or the same object at different scales) on the same input image using HOG. More...
 
class  CLHOGOrientationBinningKernel
 OpenCL kernel to perform HOG Orientation Binning. More...
 
class  CLIm2ColKernel
 Interface for the im2col reshape kernel. More...
 
class  CLIntegralImage
 Basic function to execute integral image. More...
 
class  CLIntegralImageHorKernel
 Interface to run the horizontal pass of the integral image kernel. More...
 
class  CLIntegralImageVertKernel
 Interface to run the vertical pass of the integral image kernel. More...
 
class  CLKernelLibrary
 CLKernelLibrary class. More...
 
class  CLL2NormalizeLayer
 Basic function to perform a L2 normalization on a given axis. More...
 
class  CLL2NormalizeLayerKernel
 Interface for performing a L2 normalize on a given axis given the square sum of it in this axis. More...
 
class  CLLaplacianPyramid
 Basic function to execute laplacian pyramid. More...
 
class  CLLaplacianReconstruct
 Basic function to execute laplacian reconstruction. More...
 
struct  CLLKInternalKeypoint
 Internal keypoint structure for Lucas-Kanade Optical Flow. More...
 
class  CLLKTrackerFinalizeKernel
 Interface to run the finalize step of LKTracker, where it truncates the coordinates stored in new_points array. More...
 
class  CLLKTrackerInitKernel
 Interface to run the initialization step of LKTracker. More...
 
class  CLLKTrackerStage0Kernel
 Interface to run the first stage of LKTracker, where A11, A12, A22, min_eig, ival, ixval and iyval are computed. More...
 
class  CLLKTrackerStage1Kernel
 Interface to run the second stage of LKTracker, where the motion vectors of the given points are computed. More...
 
class  CLLocallyConnectedLayer
 Basic function to compute the locally connected layer. More...
 
class  CLLocallyConnectedMatrixMultiplyKernel
 OpenCL kernel to multiply each row of first tensor with low 2 dimensions of second tensor. More...
 
class  CLLogits1DMaxKernel
 Interface for the identifying the max value of 1D Logits. More...
 
class  CLLogits1DMaxShiftExpSumKernel
 Interface for max, shifting, exponentiating and summing the logits. More...
 
class  CLLogits1DNormKernel
 Interface for calculating the final step of the Softmax Layer where each logit value is multiplied by the inverse of the sum of the logits. More...
 
class  CLLogits1DShiftExpSumKernel
 Interface for shifting, exponentiating and summing the logits. More...
 
class  CLLogLayer
 Basic function to perform elementwise log on an input tensor. More...
 
class  CLLSTMLayer
 This function performs a single time step in a Long Short-Term Memory (LSTM) layer. More...
 
class  CLLSTMLayerQuantized
 Basic function to run CLLSTMLayerQuantized. More...
 
class  CLLut
 Basic implementation of the OpenCL lut interface. More...
 
class  CLLutAllocator
 Basic implementation of a CL memory LUT allocator. More...
 
class  CLMagnitude
 Basic function to run CLMagnitudePhaseKernel. More...
 
class  CLMagnitudePhaseKernel
 Template interface for the kernel to compute magnitude and phase. More...
 
class  CLMeanStdDev
 Basic function to execute mean and standard deviation by calling CLMeanStdDevKernel. More...
 
class  CLMeanStdDevKernel
 Interface for the kernel to calculate mean and standard deviation of input image pixels. More...
 
class  CLMeanStdDevNormalizationKernel
 Interface for the kernel to normalize the input 2D tensor across the first dimension with respect to mean and standard deviation of the same dimension. More...
 
class  CLMeanStdDevNormalizationLayer
 Basic function to execute mean and standard deviation normalization by calling CLMeanStdDevNormalizationKernel. More...
 
class  CLMedian3x3
 Basic function to execute median filter. More...
 
class  CLMedian3x3Kernel
 Interface for the median 3x3 filter kernel. More...
 
class  CLMemory
 OpenCL implementation of memory object. More...
 
class  CLMemsetKernel
 Interface for filling the planes of a tensor. More...
 
class  CLMinMaxKernel
 Interface for the kernel to perform min max search on an image. More...
 
class  CLMinMaxLayerKernel
 Interface for the kernel to perform min max search on a 3D tensor. More...
 
class  CLMinMaxLocation
 Basic function to execute min and max location. More...
 
class  CLMinMaxLocationKernel
 Interface for the kernel to find min max locations of an image. More...
 
class  CLMultiHOG
 Basic implementation of the CL multi HOG data-objects. More...
 
class  CLMultiImage
 Basic implementation of the CL multi-planar image interface. More...
 
class  CLNegLayer
 Basic function to negate an input tensor. More...
 
class  CLNonLinearFilter
 Basic function to execute non linear filter. More...
 
class  CLNonLinearFilterKernel
 Interface for the kernel to apply a non-linear filter. More...
 
class  CLNonMaximaSuppression3x3
 Basic function to execute non-maxima suppression over a 3x3 window. More...
 
class  CLNonMaximaSuppression3x3Kernel
 Interface to perform Non-Maxima suppression over a 3x3 window using OpenCL. More...
 
class  CLNormalizationLayer
 Basic function to compute a normalization layer. More...
 
class  CLNormalizationLayerKernel
 Interface for the normalization layer kernel. More...
 
class  CLNormalizePlanarYUVLayer
 Basic function to run CLNormalizePlanarYUVLayerKernel. More...
 
class  CLNormalizePlanarYUVLayerKernel
 Interface for the NormalizePlanarYUV layer kernel. More...
 
struct  CLOldValue
 Structure for storing ival, ixval and iyval for each point inside the window. More...
 
class  CLOpticalFlow
 Basic function to execute optical flow. More...
 
class  CLPadLayer
 Basic function to pad a tensor. More...
 
class  CLPermute
 Basic function to execute an CLPermuteKernel. More...
 
class  CLPermuteKernel
 OpenCL kernel to perform tensor permutation. More...
 
class  CLPhase
 Basic function to execute an CLMagnitudePhaseKernel. More...
 
class  CLPixelWiseMultiplication
 Basic function to run CLPixelWiseMultiplicationKernel. More...
 
class  CLPixelWiseMultiplicationKernel
 Interface for the pixelwise multiplication kernel. More...
 
class  CLPoolingLayer
 Basic function to simulate a pooling layer with the specified pooling operation. More...
 
class  CLPoolingLayerKernel
 Interface for the pooling layer kernel. More...
 
class  CLPReluLayer
 Basic function to run CLArithmeticOperationKernel for PRELU. More...
 
class  CLPriorBoxLayer
 Basic function to run CLPriorBoxLayerKernel. More...
 
class  CLPriorBoxLayerKernel
 Interface for the PriorBox layer kernel. More...
 
class  CLPyramid
 Basic implementation of the OpenCL pyramid interface. More...
 
struct  CLQuantization
 OpenCL quantization data. More...
 
class  CLQuantizationLayer
 Basic function to simulate a quantization layer. More...
 
class  CLQuantizationLayerKernel
 Interface for the quantization layer kernel. More...
 
class  CLRange
 Basic function to run CLRangeKernel. More...
 
class  CLRangeKernel
 Kernel class for Range. More...
 
class  CLReduceMean
 Basic function to perform reduce operation. More...
 
class  CLReductionOperation
 Perform reduction operation. More...
 
class  CLReductionOperationKernel
 Interface for the reduction operation kernel. More...
 
class  CLRemap
 Basic function to execute remap. More...
 
class  CLRemapKernel
 OpenCL kernel to perform a remap on a tensor. More...
 
class  CLReorgLayer
 
class  CLReorgLayerKernel
 OpenCL kernel to perform a reorg layer. More...
 
class  CLReshapeLayer
 Basic function to run CLReshapeLayerKernel. More...
 
class  CLReshapeLayerKernel
 Interface for the kernel to perform tensor reshaping. More...
 
class  CLReverse
 Basic function to run CLReverseKernel. More...
 
class  CLReverseKernel
 Interface for the reverse kernel. More...
 
class  CLRNNLayer
 Basic function to run CLRNNLayer. More...
 
class  CLROIAlignLayer
 Basic function to run CLROIAlignLayerKernel. More...
 
class  CLROIAlignLayerKernel
 Interface for the RoIAlign kernel. More...
 
class  CLROIPoolingLayer
 Basic function to run CLROIPoolingLayerKernel. More...
 
class  CLROIPoolingLayerKernel
 Interface for the ROI pooling layer kernel. More...
 
class  CLRoundLayer
 Basic function to get the round (to the nearest even) value of an input tensor. More...
 
class  CLRsqrtLayer
 Basic function to perform inverse square root on an input tensor. More...
 
class  CLSaturatedArithmeticOperationKernel
 Addition operation. More...
 
class  CLScale
 Basic function to run CLScaleKernel. More...
 
class  CLScaleKernel
 Interface for the scale kernel. More...
 
class  CLScharr3x3
 Basic function to execute scharr 3x3 filter. More...
 
class  CLScharr3x3Kernel
 Interface for the kernel to run a 3x3 Scharr filter on a tensor. More...
 
class  CLScheduler
 Provides global access to a CL context and command queue. More...
 
class  CLSelect
 Basic function to run CLSelect. More...
 
class  CLSelectKernel
 OpenCL interface for executing the select kernel. More...
 
class  CLSeparableConvolutionHorKernel
 Kernel for the Horizontal pass of a Separable Convolution. More...
 
class  CLSeparableConvolutionVertKernel
 Kernel for the Vertical pass of a Separable Convolution. More...
 
class  CLSinLayer
 Basic function to calculate sine of an input tensor. More...
 
class  CLSlice
 Basic function to perform tensor slicing. More...
 
class  CLSobel3x3
 Basic function to execute sobel 3x3 filter. More...
 
class  CLSobel3x3Kernel
 Interface for the kernel to run a 3x3 Sobel filter on a tensor. More...
 
class  CLSobel5x5
 Basic function to execute sobel 5x5 filter. More...
 
class  CLSobel5x5HorKernel
 Interface for the kernel to run the horizontal pass of 5x5 Sobel filter on a tensor. More...
 
class  CLSobel5x5VertKernel
 Interface for the kernel to run the vertical pass of 5x5 Sobel filter on a tensor. More...
 
class  CLSobel7x7
 Basic function to execute sobel 7x7 filter. More...
 
class  CLSobel7x7HorKernel
 Interface for the kernel to run the horizontal pass of 7x7 Sobel filter on a tensor. More...
 
class  CLSobel7x7VertKernel
 Interface for the kernel to run the vertical pass of 7x7 Sobel filter on a tensor. More...
 
class  CLSoftmaxLayer
 Basic function to compute a SoftmaxLayer. More...
 
class  CLSpaceToBatchLayer
 Basic function to spatial divide a tensor. More...
 
class  CLSpaceToBatchLayerKernel
 Interface for the space to batch kernel. More...
 
class  CLSpaceToDepthLayer
 Basic function to run CLSpaceToDepthLayerKernel. More...
 
class  CLSpaceToDepthLayerKernel
 Interface for the space to depth kernel. More...
 
class  CLSplit
 Basic function to split a tensor along a given axis. More...
 
class  CLStackLayer
 Basic function to stack tensors along an axis. More...
 
class  CLStackLayerKernel
 OpenCL kernel to stacks a rank-R tensor into one with rank-(R+1) along the axis dimension. More...
 
class  CLStridedSlice
 Basic function to run CLStridedSliceKernel. More...
 
class  CLStridedSliceKernel
 Interface for the kernel to perform tensor strided slicing. More...
 
class  CLSubTensor
 Basic implementation of the OpenCL sub-tensor interface. More...
 
class  CLSymbols
 Class for loading OpenCL symbols. More...
 
class  CLTableLookup
 Basic function to run CLTableLookupKernel. More...
 
class  CLTableLookupKernel
 Interface for the kernel to perform table lookup calculations. More...
 
class  CLTensor
 Basic implementation of the OpenCL tensor interface. More...
 
class  CLTensorAllocator
 Basic implementation of a CL memory tensor allocator. More...
 
class  CLThreshold
 Basic function to run CLThresholdKernel. More...
 
class  CLThresholdKernel
 Interface for the thresholding kernel. More...
 
class  CLTile
 Basic function to run CLTileKernel. More...
 
class  CLTileKernel
 OpenCL kernel to perform a Tile operation. More...
 
class  CLTranspose
 Basic function to transpose a matrix on OpenCL. More...
 
class  CLTransposeKernel
 OpenCL kernel which transposes the elements of a matrix. More...
 
class  CLTuner
 Basic implementation of the OpenCL tuner interface. More...
 
class  CLUnstack
 Basic function to unpack a rank-R tensor into rank-(R-1) tensors. More...
 
class  CLUpsampleLayer
 Basic function to run CLUpsampleLayerKernel. More...
 
class  CLUpsampleLayerKernel
 Interface for the UpsampleLayer kernel on OpenCL. More...
 
class  CLWarpAffine
 Basic function to run CLWarpAffineKernel for AFFINE transformation. More...
 
class  CLWarpAffineKernel
 Interface for the warp affine kernel. More...
 
class  CLWarpPerspective
 Basic function to run CLWarpPerspectiveKernel for PERSPECTIVE transformation. More...
 
class  CLWarpPerspectiveKernel
 Interface for the warp perspective kernel. More...
 
class  CLWeightsReshapeKernel
 OpenCL kernel to perform reshaping on the weights used by convolution and locally connected layer. More...
 
class  CLWidthConcatenate2TensorsKernel
 Interface for the width concatenate kernel of 2 tensors. More...
 
class  CLWidthConcatenate4TensorsKernel
 Interface for the width concatenate kernel of 4 tensors. More...
 
class  CLWidthConcatenateLayerKernel
 Interface for the width concatenate kernel. More...
 
class  CLWinogradConvolutionLayer
 Basic function to execute Winograd-based convolution on OpenCL. More...
 
class  CLWinogradFilterTransformKernel
 Interface for the Winograd filter transform kernel. More...
 
class  CLWinogradInputTransform
 Basic function to execute a CLWinogradInputTransformKernel. More...
 
class  CLWinogradInputTransformKernel
 OpenCL kernel to perform Winograd input transform. More...
 
class  CLWinogradOutputTransformKernel
 Interface for the Winograd output transform kernel. More...
 
class  CLYOLOLayer
 Basic function to run CLYOLOLayerKernel that performs a partial activation on the input. More...
 
class  CLYOLOLayerKernel
 Interface for the YOLO layer kernel that performs partial activation. More...
 
class  ComputeAnchorsInfo
 ComputeAnchors information class. More...
 
class  Coordinates
 Coordinates of an item. More...
 
struct  Coordinates2D
 Coordinate type. More...
 
struct  Coordinates3D
 Coordinate type. More...
 
class  CPPBoxWithNonMaximaSuppressionLimit
 Basic function to run CPPBoxWithNonMaximaSuppressionLimitKernel. More...
 
class  CPPBoxWithNonMaximaSuppressionLimitKernel
 CPP kernel to perform computation of BoxWithNonMaximaSuppressionLimit. More...
 
class  CPPCornerCandidatesKernel
 CPP kernel to perform corner candidates. More...
 
class  CPPDetectionOutputLayer
 CPP Function to generate the detection output based on location and confidence predictions by doing non maximum suppression. More...
 
class  CPPDetectionPostProcessLayer
 CPP Function to generate the detection output based on center size encoded boxes, class prediction and anchors by doing non maximum suppression. More...
 
class  CPPDetectionWindowNonMaximaSuppressionKernel
 CPP kernel to perform in-place computation of euclidean distance on IDetectionWindowArray. More...
 
class  CPPFlipWeightsKernel
 CPP kernel to perform 180 degrees flipping on deconvolution weights. More...
 
class  CPPNonMaximumSuppression
 CPP Function to perform non maximum suppression on the bounding boxes and scores. More...
 
class  CPPNonMaximumSuppressionKernel
 CPP Function to perform non maximum suppression on the bounding boxes and scores. More...
 
class  CPPPermute
 Basic function to run CPPPermuteKernel. More...
 
class  CPPPermuteKernel
 CPP kernel to perform tensor permutation. More...
 
class  CPPScheduler
 C++11 implementation of a pool of threads to automatically split a kernel's execution among several threads. More...
 
class  CPPSortEuclideanDistanceKernel
 CPP kernel to perform sorting and euclidean distance. More...
 
class  CPPTopKV
 Basic function to run CPPTopKVKernel. More...
 
class  CPPTopKVKernel
 CPP kernel to perform tensor TopKV operation. More...
 
class  CPPUpsample
 Basic function to run CPPUpsample. More...
 
class  CPPUpsampleKernel
 CPP kernel to perform tensor upsample. More...
 
class  CPUInfo
 
struct  DepthwiseConvolutionReshapeInfo
 
class  DetectionOutputLayerInfo
 Detection Output layer info. More...
 
class  DetectionPostProcessLayerInfo
 Detection Output layer info. More...
 
struct  DetectionWindow
 Detection window used for the object detection. More...
 
class  Dimensions
 Dimensions with dimensionality. More...
 
class  Distribution1D
 Basic implementation of the 1D distribution interface. More...
 
struct  enable_bitwise_ops
 Disable bitwise operations by default. More...
 
struct  enable_bitwise_ops< arm_compute::GPUTarget >
 Enable bitwise operations on GPUTarget enumerations. More...
 
struct  FFT1DInfo
 Descriptor used by the FFT1D function. More...
 
struct  FFT2DInfo
 Descriptor used by the FFT2D function. More...
 
struct  FFTDigitReverseKernelInfo
 Descriptor for FFT digit reverse kernels. More...
 
struct  FFTRadixStageKernelInfo
 Descriptor used by the FFT core kernels. More...
 
struct  FFTScaleKernelInfo
 Descriptor for FFT scale kernels. More...
 
struct  FullyConnectedLayerInfo
 Fully connected layer info. More...
 
class  GCAbsoluteDifference
 Basic function to run GCAbsoluteDifferenceKernel. More...
 
class  GCAbsoluteDifferenceKernel
 Interface for the absolute difference kernel. More...
 
class  GCActivationLayer
 Basic function to run GCActivationLayerKernel. More...
 
class  GCActivationLayerKernel
 Interface for the activation layer kernel. More...
 
class  GCArithmeticAddition
 Basic function to run GCArithmeticAdditionKernel. More...
 
class  GCArithmeticAdditionKernel
 Interface for the arithmetic addition kernel. More...
 
class  GCBatchNormalizationLayer
 Basic function to run GCBatchNormalizationLayerKernel and simulate a batch normalization layer. More...
 
class  GCBatchNormalizationLayerKernel
 Interface for the BatchNormalization layer kernel. More...
 
class  GCBufferAllocator
 Default GLES buffer allocator implementation. More...
 
class  GCBufferMemoryRegion
 GLES buffer memory region implementation. More...
 
class  GCCol2ImKernel
 Interface for the col2im reshaping kernel. More...
 
class  GCConcatenateLayer
 Basic function to execute concatenate tensors along a given axis. More...
 
class  GCConvolutionLayer
 Basic function to compute the convolution layer. More...
 
class  GCConvolutionLayerReshapeWeights
 Function to reshape and transpose the weights. More...
 
class  GCDepthConcatenateLayerKernel
 Interface for the depth concatenate kernel. More...
 
class  GCDepthwiseConvolutionLayer3x3
 Basic function to execute a depthwise convolution for kernel size 3x3xC. More...
 
class  GCDepthwiseConvolutionLayer3x3Kernel
 Interface for the kernel to run a 3x3 depthwise convolution on a tensor. More...
 
class  GCDirectConvolutionLayer
 Basic function to execute direct convolution function. More...
 
class  GCDirectConvolutionLayerKernel
 Interface for the direct convolution kernel. More...
 
class  GCDropoutLayer
 Basic function to do dropout op. More...
 
class  GCDropoutLayerKernel
 Interface for the dropout layer kernel. More...
 
class  GCFillBorder
 Basic function to run GCFillBorderKernel. More...
 
class  GCFillBorderKernel
 Interface for filling the border of a kernel. More...
 
class  GCFullyConnectedLayer
 Basic function to compute a Fully Connected layer on OpenGL ES. More...
 
class  GCFullyConnectedLayerReshapeWeights
 Basic function to reshape the weights of Fully Connected layer with OpenGL ES. More...
 
class  GCGEMM
 Basic function to execute GEMM on OpenGLES Compute. More...
 
class  GCGEMMInterleave4x4
 Basic function to execute GCGEMMInterleave4x4Kernel. More...
 
class  GCGEMMInterleave4x4Kernel
 OpenGL ES kernel which interleaves the elements of a matrix A in chunk of 4x4. More...
 
class  GCGEMMMatrixAccumulateBiasesKernel
 Interface to add a bias to each row of the input tensor. More...
 
class  GCGEMMMatrixAdditionKernel
 OpenGL ES kernel to perform the in-place matrix addition between 2 matrices, taking into account that the second matrix might be weighted by a scalar value beta. More...
 
class  GCGEMMMatrixMultiplyKernel
 GLES Compute kernel to multiply two input matrices "A" and "B" or to multiply a vector "A" by a matrix "B". More...
 
class  GCGEMMTranspose1xW
 Basic function to execute GCGEMMTranspose1xWKernel. More...
 
class  GCGEMMTranspose1xWKernel
 OpenGLES kernel which transposes the elements of a matrix in chunks of 1xW, where W is equal to (16 / element size of the tensor) More...
 
class  GCIm2ColKernel
 Interface for the im2col reshape kernel. More...
 
class  GCKernel
 GCKernel class. More...
 
class  GCKernelLibrary
 GCKernelLibrary class. More...
 
class  GCLogits1DMaxKernel
 Interface for the identifying the max value of 1D Logits. More...
 
class  GCLogits1DNormKernel
 Interface for calculating the final step of the Softmax Layer where each logit value is multiplied by the inverse of the sum of the logits. More...
 
class  GCLogits1DShiftExpSumKernel
 Interface for shifting the logits values around the max value and exponentiating the result. More...
 
class  GCMemory
 GLES implementation of memory object. More...
 
class  GCNormalizationLayer
 Basic function to compute a normalization layer. More...
 
class  GCNormalizationLayerKernel
 Interface for the normalization layer kernel. More...
 
class  GCNormalizePlanarYUVLayer
 Basic function to run GCNormalizePlanarYUVLayerKernel. More...
 
class  GCNormalizePlanarYUVLayerKernel
 Interface for the NormalizePlanarYUV layer kernel. More...
 
class  GCPixelWiseMultiplication
 Basic function to run GCPixelWiseMultiplicationKernel. More...
 
class  GCPixelWiseMultiplicationKernel
 Interface for the pixelwise multiplication kernel. More...
 
class  GCPoolingLayer
 Basic function to simulate a pooling layer with the specified pooling operation. More...
 
class  GCPoolingLayerKernel
 Interface for the pooling layer kernel. More...
 
class  GCProgram
 GCProgram class. More...
 
class  GCScale
 Basic function to run GCScaleKernel. More...
 
class  GCScaleKernel
 Interface for the scale kernel. More...
 
class  GCScheduler
 Provides global access to a OpenGL ES context and command queue. More...
 
class  GCSoftmaxLayer
 Basic function to compute a SoftmaxLayer. More...
 
class  GCTensor
 Interface for OpenGL ES tensor. More...
 
class  GCTensorAllocator
 Basic implementation of a GLES memory tensor allocator. More...
 
class  GCTensorShift
 Basic function to execute shift function for tensor. More...
 
class  GCTensorShiftKernel
 Interface for the kernel to shift valid data on a tensor. More...
 
class  GCTranspose
 Basic function to transpose a matrix on OpenGL ES. More...
 
class  GCTransposeKernel
 OpenGL ES kernel which transposes the elements of a matrix. More...
 
class  GCWeightsReshapeKernel
 GLES Compute kernel to perform reshaping on the weights used by convolution and locally connected layer. More...
 
class  GEMMInfo
 GEMM information class. More...
 
struct  GEMMKernelInfo
 Descriptor used by the GEMM kernels. More...
 
struct  GEMMLHSMatrixInfo
 GEMM LHS (Left Hand Side) matrix information. More...
 
struct  GEMMLowpOutputStageInfo
 GEMMLowp output stage info. More...
 
class  GEMMReshapeInfo
 GEMM reshape information class. More...
 
struct  GEMMRHSMatrixInfo
 GEMM RHS (Right Hand Side) matrix information. More...
 
class  GenerateProposalsInfo
 Generate Proposals Information class. More...
 
class  HOG
 CPU implementation of HOG data-object. More...
 
class  HOGInfo
 Store the HOG's metadata. More...
 
class  IAccessWindow
 Interface describing methods to update access window and padding based on kernel parameters. More...
 
class  IAllocator
 Allocator interface. More...
 
class  IArray
 Array of type T. More...
 
class  IBufferManager
 Buffer manager used when reshaping B on the fly. More...
 
class  ICLArray
 Interface for OpenCL Array. More...
 
class  ICLDepthwiseConvolutionLayer3x3Kernel
 Interface for the kernel to run a 3x3 depthwise convolution on a tensor. More...
 
class  ICLDistribution1D
 ICLDistribution1D interface class. More...
 
class  ICLGEMMKernelConfiguration
 Basic interface for the GEMM kernel configuration. More...
 
class  ICLGEMMLowpReductionKernel
 Common interface for all OpenCL reduction kernels. More...
 
class  ICLHOG
 Interface for OpenCL HOG data-object. More...
 
class  ICLKernel
 Common interface for all the OpenCL kernels. More...
 
class  ICLLut
 Interface for OpenCL LUT. More...
 
class  ICLMemoryRegion
 OpenCL memory region interface. More...
 
class  ICLMultiHOG
 Interface for storing multiple HOG data-objects. More...
 
class  ICLMultiImage
 Interface for OpenCL multi-planar images. More...
 
class  ICLSimple2DKernel
 Interface for simple OpenCL kernels having 1 tensor input and 1 tensor output. More...
 
class  ICLSimple3DKernel
 Interface for simple OpenCL kernels having 1 tensor input and 1 tensor output. More...
 
class  ICLSimpleFunction
 Basic interface for functions which have a single OpenCL kernel. More...
 
class  ICLSimpleKernel
 Interface for simple OpenCL kernels having 1 tensor input and 1 tensor output. More...
 
class  ICLSVMMemoryRegion
 OpenCL SVM memory region interface. More...
 
class  ICLTensor
 Interface for OpenCL tensor. More...
 
class  ICLTuner
 Basic interface for tuning the OpenCL kernels. More...
 
class  ICPPKernel
 Common interface for all kernels implemented in C++. More...
 
class  ICPPSimpleFunction
 Basic interface for functions which have a single CPP kernel. More...
 
class  ICPPSimpleKernel
 Interface for simple C++ kernels having 1 tensor input and 1 tensor output. More...
 
class  IDistribution
 Interface for distribution objects. More...
 
class  IDistribution1D
 1D Distribution interface More...
 
class  IFunction
 Base class for all functions. More...
 
class  IGCKernel
 Common interface for all the GLES kernels. More...
 
class  IGCMemoryRegion
 GLES memory region interface. More...
 
class  IGCSimple2DKernel
 Interface for simple OpenGL ES kernels having 1 tensor input and 1 tensor output. More...
 
class  IGCSimple3DKernel
 Interface for simple GLES kernels having 1 tensor input and 1 tensor output. More...
 
class  IGCSimpleFunction
 Basic interface for functions which have a single OpenGL ES kernel. More...
 
class  IGCSimpleKernel
 Interface for simple OpenGL ES kernels having 1 tensor input and 1 tensor output. More...
 
class  IGCTensor
 Interface for GLES Compute tensor. More...
 
class  IHOG
 Interface for HOG data-object. More...
 
class  IKernel
 Common information for all the kernels. More...
 
class  ILifetimeManager
 Interface for managing the lifetime of objects. More...
 
class  ILut
 Lookup Table object interface. More...
 
class  ILutAllocator
 Basic interface to allocate LUTs'. More...
 
class  IMemory
 Memory interface. More...
 
class  IMemoryGroup
 Memory group interface. More...
 
class  IMemoryManager
 Memory manager interface to handle allocations of backing memory. More...
 
class  IMemoryPool
 Memory Pool Inteface. More...
 
class  IMemoryRegion
 Memory region interface. More...
 
class  IMultiHOG
 Interface for storing multiple HOG data-objects. More...
 
class  IMultiImage
 Interface for multi-planar images. More...
 
class  INEGEMMLowpReductionKernel
 Common interface for all NEON reduction kernels. More...
 
class  INEHarrisScoreKernel
 Common interface for all Harris Score kernels. More...
 
class  INESimpleFunction
 Basic interface for functions which have a single NEON kernel. More...
 
class  INESimpleFunctionNoBorder
 Basic interface for functions which have a single NEON kernel and no border. More...
 
class  INEWarpKernel
 Common interface for warp affine and warp perspective. More...
 
class  INEWinogradLayerTransformInputKernel
 Interface for the NEON kernel to perform Winograd input transform. More...
 
class  INEWinogradLayerTransformOutputKernel
 Interface for the NEON kernel to perform Winograd output transform. More...
 
class  INEWinogradLayerTransformWeightsKernel
 Interface for the NEON kernel to perform Winograd weights transform. More...
 
struct  InternalKeyPoint
 Internal keypoint class for Lucas-Kanade Optical Flow. More...
 
struct  IOFormatInfo
 IO formatting information class. More...
 
class  IPoolManager
 Memory pool manager interface. More...
 
class  IPyramid
 Interface for pyramid data-object. More...
 
class  IScheduler
 Scheduler interface to run kernels. More...
 
class  ISimpleLifetimeManager
 Abstract class of the simple lifetime manager interface. More...
 
class  ITensor
 Interface for NEON tensor. More...
 
class  ITensorAllocator
 Interface to allocate tensors. More...
 
class  ITensorInfo
 Store the tensor's metadata. More...
 
class  Iterator
 Iterator updated by execute_window_loop for each window element. More...
 
class  Kernel
 Kernel class. More...
 
struct  KeyPoint
 Keypoint type. More...
 
class  LSTMParams
 
class  Lut
 Basic implementation of the LUT interface. More...
 
class  LutAllocator
 Basic implementation of a CPU memory LUT allocator. More...
 
class  MEMInfo
 
class  Memory
 CPU implementation of memory object. More...
 
class  MemoryGroupBase
 Memory group. More...
 
class  MemoryGroupResourceScope
 Memory group resources scope handling class. More...
 
class  MemoryManagerOnDemand
 On-demand memory manager. More...
 
class  MemoryRegion
 Memory region CPU implementation. More...
 
struct  MinMaxLocationValues
 Min and max values and locations. More...
 
class  MultiHOG
 CPU implementation of multi HOG data-object. More...
 
class  MultiImage
 Basic implementation of the multi-planar image interface. More...
 
class  MultiImageInfo
 Store the multi-planar image's metadata. More...
 
class  NEAbsLayer
 Basic function to compute the absolute value of an input tensor. More...
 
class  NEAbsoluteDifference
 Basic function to run NEAbsoluteDifferenceKernel. More...
 
class  NEAbsoluteDifferenceKernel
 Interface for the absolute difference kernel. More...
 
class  NEAccumulate
 Basic function to run NEAccumulateKernel. More...
 
class  NEAccumulateKernel
 Interface for the accumulate kernel. More...
 
class  NEAccumulateSquared
 Basic function to run NEAccumulateSquaredKernel. More...
 
class  NEAccumulateSquaredKernel
 Interface for the accumulate squared kernel. More...
 
class  NEAccumulateWeighted
 Basic function to run NEAccumulateWeightedKernel. More...
 
class  NEAccumulateWeightedKernel
 Interface for the accumulate weighted kernel. More...
 
class  NEActivationLayer
 Basic function to run NEActivationLayerKernel. More...
 
class  NEActivationLayerKernel
 Interface for the activation layer kernel. More...
 
class  NEArgMinMaxLayer
 Function to calculate the index of the minimum or maximum values in a tensor based on an axis. More...
 
class  NEArithmeticAddition
 Basic function to run NEArithmeticAdditionKernel. More...
 
class  NEArithmeticAdditionKernel
 Interface for the kernel to perform addition between two tensors. More...
 
class  NEArithmeticOperationKernel
 
class  NEArithmeticSubtraction
 Basic function to run NEArithmeticSubtractionKernel. More...
 
class  NEArithmeticSubtractionKernel
 Interface for the kernel to perform subtraction between two tensors. More...
 
class  NEBatchConcatenateLayerKernel
 Interface for the batch concatenate kernel. More...
 
class  NEBatchNormalizationLayer
 Basic function to run NENormalizationLayerKernel and simulate a batch normalization layer. More...
 
class  NEBatchNormalizationLayerKernel
 Interface for the batch normalization layer kernel. More...
 
class  NEBatchToSpaceLayer
 Basic function to run NEBatchToSpaceLayerKernel. More...
 
class  NEBatchToSpaceLayerKernel
 Interface for the batch to space kernel. More...
 
class  NEBitwiseAnd
 Basic function to run NEBitwiseAndKernel. More...
 
class  NEBitwiseAndKernel
 Interface for the kernel to perform bitwise AND between XY-planes of two tensors. More...
 
class  NEBitwiseNot
 Basic function to run NEBitwiseNotKernel. More...
 
class  NEBitwiseNotKernel
 Interface for the kernel to perform bitwise NOT operation. More...
 
class  NEBitwiseOr
 Basic function to run NEBitwiseOrKernel. More...
 
class  NEBitwiseOrKernel
 Interface for the kernel to perform bitwise inclusive OR between two tensors. More...
 
class  NEBitwiseXor
 Basic function to run NEBitwiseXorKernel. More...
 
class  NEBitwiseXorKernel
 Interface for the kernel to perform bitwise exclusive OR (XOR) between two tensors. More...
 
class  NEBox3x3
 Basic function to execute box filter 3x3. More...
 
class  NEBox3x3Kernel
 NEON kernel to perform a Box 3x3 filter. More...
 
class  NECannyEdge
 Basic function to execute canny edge on NEON. More...
 
class  NECast
 Basic function to run NEDepthConvertLayerKernel. More...
 
class  NEChannelCombine
 Basic function to run NEChannelCombineKernel to perform channel combination. More...
 
class  NEChannelCombineKernel
 Interface for the channel combine kernel. More...
 
class  NEChannelExtract
 Basic function to run NEChannelExtractKernel to perform channel extraction. More...
 
class  NEChannelExtractKernel
 Interface for the channel extract kernel. More...
 
class  NEChannelShuffleLayer
 Basic function to run NEChannelShuffleLayerKernel. More...
 
class  NEChannelShuffleLayerKernel
 Interface for the channel shuffle kernel. More...
 
class  NECol2Im
 Basic function to run NECol2Im. More...
 
class  NECol2ImKernel
 NEON kernel to perform col2im reshaping. More...
 
class  NEColorConvert
 Basic function to run NEColorConvertKernel to perform color conversion. More...
 
class  NEColorConvertKernel
 Interface for the color convert kernel. More...
 
class  NEComparisonOperationKernel
 
class  NEComplexPixelWiseMultiplication
 Basic function to run NEComplexPixelWiseMultiplicationKernel. More...
 
class  NEComplexPixelWiseMultiplicationKernel
 Interface for the complex pixelwise multiplication kernel. More...
 
class  NEConcatenateLayer
 Basic function to execute concatenate tensors along a given axis. More...
 
class  NEConvertFullyConnectedWeights
 Basic function to run NEConvertFullyConnectedWeightsKernel. More...
 
class  NEConvertFullyConnectedWeightsKernel
 Interface to convert the 2D Fully Connected weights from NCHW to NHWC or vice versa. More...
 
class  NEConvolution3x3
 Basic function to execute convolution of size 3x3. More...
 
class  NEConvolutionKernel
 Interface for the kernel to run an arbitrary size convolution on a tensor. More...
 
class  NEConvolutionLayer
 Basic function to simulate a convolution layer. More...
 
class  NEConvolutionLayerReshapeWeights
 Function to reshape the weights. More...
 
class  NEConvolutionRectangle
 Basic function to execute non-square convolution. More...
 
class  NEConvolutionRectangleKernel
 Kernel for the running convolution on a rectangle matrix. More...
 
class  NEConvolutionSquare
 Basic function to execute convolution of size 5x5, 7x7, 9x9. More...
 
class  NECopy
 Basic function to run NECopyKernel. More...
 
class  NECopyKernel
 NEON kernel to perform a copy between two tensors. More...
 
class  NECropKernel
 Interface for the kernel to perform tensor cropping. More...
 
class  NECropResize
 Function to perform cropping and resizing. More...
 
class  NECumulativeDistributionKernel
 Interface for the cumulative distribution (cummulative summmation) calculation kernel. More...
 
class  NEDeconvolutionLayer
 Function to run the deconvolution layer. More...
 
class  NEDepthConcatenateLayerKernel
 Interface for the depth concatenate kernel. More...
 
class  NEDepthConvertLayer
 Basic function to run NEDepthConvertLayerKernel. More...
 
class  NEDepthConvertLayerKernel
 Depth conversion kernel This function ignores the scale and zeroPoint of quanized tensors, i.e. More...
 
class  NEDepthToSpaceLayer
 Basic function to run NEDepthToSpaceLayerKernel. More...
 
class  NEDepthToSpaceLayerKernel
 Interface for the depth to space kernel. More...
 
class  NEDepthwiseConvolutionAssemblyDispatch
 Depthwise convolution assembly kernel glue. More...
 
class  NEDepthwiseConvolutionLayer
 Basic function to execute a generic depthwise convolution. More...
 
class  NEDepthwiseConvolutionLayer3x3
 Basic function to execute a depthwise convolution for kernel size 3x3xC. More...
 
class  NEDepthwiseConvolutionLayer3x3Kernel
 Interface for the kernel to run a 3x3 depthwise convolution on a tensor. More...
 
class  NEDepthwiseConvolutionLayerNativeKernel
 Interface for the kernel to run a depthwise convolution native on a tensor. More...
 
class  NEDepthwiseConvolutionLayerOptimized
 Basic function to execute optimized depthwise convolution routines. More...
 
class  NEDepthwiseIm2ColKernel
 Interface for the depthwise im2col reshape kernel. More...
 
class  NEDepthwiseSeparableConvolutionLayer
 Basic function to execute depthwise convolution. More...
 
class  NEDepthwiseVectorToTensorKernel
 Interface for the depthwise vector to tensor kernel. More...
 
class  NEDepthwiseWeightsReshapeKernel
 Interface for the depthwise weights reshape kernel. More...
 
class  NEDequantizationLayer
 Basic function to run NEDequantizationLayerKernel that dequantizes an input tensor. More...
 
class  NEDequantizationLayerKernel
 Interface for the dequantization layer kernel. More...
 
class  NEDerivative
 Basic function to execute first order derivative operator. More...
 
class  NEDerivativeKernel
 Interface for the kernel to run the derivative along the X/Y directions on a tensor. More...
 
class  NEDilate
 Basic function to execute dilate. More...
 
class  NEDilateKernel
 Interface for the kernel to perform boolean image dilatation. More...
 
class  NEDirectConvolutionLayer
 Function to run the direct convolution. More...
 
class  NEDirectConvolutionLayerKernel
 NEON interface for Direct Convolution Layer kernel. More...
 
class  NEDirectConvolutionLayerOutputStageKernel
 NEON kernel to accumulate the biases, if provided, or downscale in case of quantized input. More...
 
class  NEDivisionOperationKernel
 
class  NEEdgeNonMaxSuppressionKernel
 NEON kernel to perform Non-Maxima suppression for Canny Edge. More...
 
class  NEEdgeTraceKernel
 NEON kernel to perform Edge tracing. More...
 
class  NEElementwiseComparison
 Basic function to run NEComparisonOperationKernel. More...
 
class  NEElementwiseComparisonStatic
 Basic function to run NEComparisonOperationKernel. More...
 
class  NEElementwiseDivision
 Basic function to run NEArithmeticOperationKernel for division. More...
 
class  NEElementwiseMax
 Basic function to run NEArithmeticOperationKernel for max. More...
 
class  NEElementwiseMin
 Basic function to run NEArithmeticOperationKernel for min. More...
 
class  NEElementwiseOperationKernel
 Interface for an element-wise operation kernel. More...
 
class  NEElementwisePower
 Basic function to run NEArithmeticOperationKernel for power. More...
 
class  NEElementwiseSquaredDiff
 Basic function to run NEArithmeticOperationKernel for squared difference. More...
 
class  NEElementwiseUnaryKernel
 Interface for an element-wise unary operation kernel. More...
 
class  NEEqualizeHistogram
 Basic function to execute histogram equalization. More...
 
class  NEErode
 Basic function to execute erode. More...
 
class  NEErodeKernel
 Interface for the kernel to perform boolean image erosion. More...
 
class  NEExpLayer
 Basic function to perform exponential on an input tensor. More...
 
class  NEFastCorners
 Basic function to execute fast corners. More...
 
class  NEFastCornersKernel
 NEON kernel to perform fast corners. More...
 
class  NEFFT1D
 Basic function to execute one dimensional FFT. More...
 
class  NEFFT2D
 Basic function to execute two dimensional FFT. More...
 
class  NEFFTConvolutionLayer
 Basic function to execute FFT-based convolution on NEON. More...
 
class  NEFFTDigitReverseKernel
 Interface for the digit reverse operation kernel. More...
 
class  NEFFTRadixStageKernel
 Interface for the FFT kernel. More...
 
class  NEFFTScaleKernel
 Interface for the inverse fft scale kernel. More...
 
class  NEFillArrayKernel
 This kernel adds all texels greater than or equal to the threshold value to the keypoint array. More...
 
class  NEFillBorder
 Basic function to run NEFillBorderKernel. More...
 
class  NEFillBorderKernel
 Interface for the kernel to fill borders. More...
 
class  NEFillInnerBorderKernel
 Interface for the kernel to fill the interior borders. More...
 
class  NEFlattenLayer
 Basic function to execute flatten layer kernel. More...
 
class  NEFlattenLayerKernel
 Interface for the flatten layer kernel. More...
 
class  NEFloor
 Basic function to run NEFloorKernel. More...
 
class  NEFloorKernel
 NEON kernel to perform a floor operation. More...
 
class  NEFullyConnectedLayer
 Basic function to compute a Fully Connected layer on NEON. More...
 
class  NEFullyConnectedLayerReshapeWeights
 Basic function to reshape the weights of Fully Connected layer with NEON. More...
 
class  NEFuseBatchNormalization
 Basic function to fuse the batch normalization node to a preceding convolution node. More...
 
class  NEFuseBatchNormalizationKernel
 OpenNE kernel to fuse the batch normalization node to a preceding convolution node. More...
 
class  NEGather
 Basic function to run NEGatherKernel. More...
 
class  NEGatherKernel
 Kernel to perform other operation on NEON. More...
 
class  NEGaussian3x3
 Basic function to execute gaussian filter 3x3. More...
 
class  NEGaussian3x3Kernel
 NEON kernel to perform a Gaussian 3x3 filter. More...
 
class  NEGaussian5x5
 Basic function to execute gaussian filter 5x5. More...
 
class  NEGaussian5x5HorKernel
 NEON kernel to perform a Gaussian 5x5 filter (horizontal pass) More...
 
class  NEGaussian5x5VertKernel
 NEON kernel to perform a Gaussian 5x5 filter (vertical pass) More...
 
class  NEGaussianPyramid
 Common interface for all Gaussian pyramid functions. More...
 
class  NEGaussianPyramidHalf
 Basic function to execute gaussian pyramid with HALF scale factor. More...
 
class  NEGaussianPyramidHorKernel
 NEON kernel to perform a GaussianPyramid (horizontal pass) More...
 
class  NEGaussianPyramidOrb
 Basic function to execute gaussian pyramid with ORB scale factor. More...
 
class  NEGaussianPyramidVertKernel
 NEON kernel to perform a GaussianPyramid (vertical pass) More...
 
class  NEGEMM
 Basic function to execute GEMM on NEON. More...
 
class  NEGEMMAssemblyBaseKernel
 Base class for GEMM NEON kernels implemented in Assembly. More...
 
class  NEGEMMAssemblyDispatch
 Assembly kernel glue. More...
 
class  NEGEMMConvolutionLayer
 Basic function to compute the convolution layer. More...
 
class  NEGEMMInterleave4x4
 Basic function to execute NEGEMMInterleave4x4Kernel. More...
 
class  NEGEMMInterleave4x4Kernel
 NEON kernel to interleave the elements of a matrix. More...
 
class  NEGEMMInterleavedWrapper
 Equivalent to arm_gemm::GemmInterleaved but using Compute Library types. More...
 
class  NEGEMMLowpAssemblyMatrixMultiplyCore
 Basic function to execute matrix multiply assembly kernels. More...
 
class  NEGEMMLowpMatrixAReductionKernel
 NEON kernel used to compute the row-vectors of sums of all the entries in each row of Matrix A. More...
 
class  NEGEMMLowpMatrixBReductionKernel
 NEON kernel used to compute the row-vectors of sums of all the entries in each column of Matrix B. More...
 
class  NEGEMMLowpMatrixMultiplyCore
 Basic function to execute GEMMLowpMatrixMultiplyCore on NEON. More...
 
class  NEGEMMLowpMatrixMultiplyKernel
 NEON kernel to multiply matrices. More...
 
class  NEGEMMLowpOffsetContributionKernel
 NEON kernel used to add the offset contribution after NEGEMMLowpMatrixMultiplyKernel. More...
 
class  NEGEMMLowpOffsetContributionOutputStageKernel
 NEON kernel used to add the offset contribution and perform the output stage after NEGEMMLowpMatrixMultiplyKernel. More...
 
class  NEGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPoint
 Basic function to execute NEGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPoint on NEON. More...
 
class  NEGEMMLowpQuantizeDownInt32ToInt16ScaleByFixedPointKernel
 NEON kernel used to quantize down the int32 accumulator values of GEMMLowp to QSYMM16. More...
 
class  NEGEMMLowpQuantizeDownInt32ToUint8Scale
 Basic function to execute NEGEMMLowpQuantizeDownInt32ToUint8Scale on NEON. More...
 
class  NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint
 Basic function to execute NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPoint on NEON. More...
 
class  NEGEMMLowpQuantizeDownInt32ToUint8ScaleByFixedPointKernel
 NEON kernel used to quantize down the int32 accumulator values of GEMMLowp to QASYMM8. More...
 
class  NEGEMMLowpQuantizeDownInt32ToUint8ScaleKernel
 NEON kernel used to quantize down the int32 accumulator values of GEMMLowp to QASYMM8. More...
 
class  NEGEMMMatrixAccumulateBiasesKernel
 NEON kernel to add a bias to each row of the input tensor. More...
 
class  NEGEMMMatrixAdditionKernel
 NEON kernel to perform the in-place matrix addition between 2 matrices taking into account that the second matrix might be weighted by a scalar value beta: More...
 
class  NEGEMMMatrixMultiplyKernel
 NEON kernel to multiply two input matrices "A" and "B". More...
 
class  NEGEMMMatrixVectorMultiplyKernel
 Interface for the GEMM matrix vector multiply kernel. More...
 
class  NEGEMMTranspose1xW
 Basic function to execute NEGEMMTranspose1xWKernel. More...
 
class  NEGEMMTranspose1xWKernel
 NEON kernel which transposes the elements of a matrix in chunks of 1xW, where W is equal to (16 / element size of the tensor) More...
 
class  NEGradientKernel
 Computes magnitude and quantised phase from inputs gradients. More...
 
class  NEHarrisCorners
 Basic function to execute harris corners detection. More...
 
class  NEHarrisScoreKernel
 Template NEON kernel to perform Harris Score. More...
 
class  NEHeightConcatenateLayerKernel
 Interface for the height concatenate kernel. More...
 
class  NEHistogram
 Basic function to run NEHistogramKernel. More...
 
class  NEHistogramKernel
 Interface for the histogram kernel. More...
 
class  NEHOGBlockNormalizationKernel
 NEON kernel to perform HOG block normalization. More...
 
class  NEHOGDescriptor
 Basic function to calculate HOG descriptor. More...
 
class  NEHOGDetector
 Basic function to execute HOG detector based on linear SVM. More...
 
class  NEHOGDetectorKernel
 NEON kernel to perform HOG detector kernel using linear SVM. More...
 
class  NEHOGGradient
 Basic function to calculate the gradient for HOG. More...
 
class  NEHOGMultiDetection
 Basic function to detect multiple objects (or the same object at different scales) on the same input image using HOG. More...
 
class  NEHOGOrientationBinningKernel
 NEON kernel to perform HOG Orientation Binning. More...
 
class  NEIm2Col
 Basic function to run NEIm2ColKernel. More...
 
class  NEIm2ColKernel
 Interface for the im2col reshape kernel. More...
 
class  NEIntegralImage
 Basic function to run a NEIntegralImageKernel. More...
 
class  NEIntegralImageKernel
 Kernel to perform an image integral on an image. More...
 
class  NEL2NormalizeLayer
 Basic function to perform a L2 normalization on a given axis. More...
 
class  NEL2NormalizeLayerKernel
 Interface for performing a L2 normalize on a given axis given the square sum of it in this axis. More...
 
class  NELaplacianPyramid
 Basic function to execute laplacian pyramid. More...
 
class  NELaplacianReconstruct
 Basic function to execute laplacian reconstruction. More...
 
struct  NELKInternalKeypoint
 Internal keypoint class for Lucas-Kanade Optical Flow. More...
 
class  NELKTrackerKernel
 Interface for the Lucas-Kanade tracker kernel. More...
 
class  NELocallyConnectedLayer
 Basic function to compute the locally connected layer. More...
 
class  NELocallyConnectedMatrixMultiplyKernel
 NEON kernel to multiply each row of first tensor with low 2 dimensions of second tensor. More...
 
class  NELogits1DMaxKernel
 Interface for the identifying the max value of 1D Logits. More...
 
class  NELogits1DSoftmaxKernel
 Interface for softmax computation for QASYMM8 with pre-computed max. More...
 
class  NELogLayer
 Basic function to compute the natural logarithm of an input tensor. More...
 
class  NELSTMLayer
 Basic function to run NELSTMLayer. More...
 
class  NELSTMLayerQuantized
 Basic function to run NELSTMLayerQuantized. More...
 
class  NEMagnitude
 Basic function to run NEMagnitudePhaseKernel. More...
 
class  NEMagnitudePhaseKernel
 Template interface for the kernel to compute magnitude and phase. More...
 
class  NEMeanStdDev
 Basic function to execute mean and std deviation. More...
 
class  NEMeanStdDevKernel
 Interface for the kernel to calculate mean and standard deviation of input image pixels. More...
 
class  NEMeanStdDevNormalizationKernel
 Interface for the kernel to normalize the input 2D tensor across the first dimension with respect to mean and standard deviation of the same dimension. More...
 
class  NEMeanStdDevNormalizationLayer
 Basic function to execute mean and standard deviation normalization by calling NEMeanStdDevNormalizationKernel. More...
 
class  NEMedian3x3
 Basic function to execute median filter. More...
 
class  NEMedian3x3Kernel
 Kernel to perform a median filter on a tensor. More...
 
class  NEMemsetKernel
 Interface for filling the planes of a tensor. More...
 
class  NEMinMaxKernel
 Interface for the kernel to perform min max search on an image. More...
 
class  NEMinMaxLayerKernel
 Interface for the kernel to perform min max search on a 3D tensor. More...
 
class  NEMinMaxLocation
 Basic function to execute min and max location. More...
 
class  NEMinMaxLocationKernel
 Interface for the kernel to find min max locations of an image. More...
 
class  NENegLayer
 Basic function to negate an input tensor. More...
 
class  NENonLinearFilter
 Basic function to execute non linear filter. More...
 
class  NENonLinearFilterKernel
 Interface for the kernel to apply a non-linear filter. More...
 
class  NENonMaximaSuppression3x3
 Basic function to execute non-maxima suppression over a 3x3 window. More...
 
class  NENonMaximaSuppression3x3Kernel
 Interface to perform Non-Maxima suppression over a 3x3 window using NEON. More...
 
class  NENormalizationLayer
 Basic function to compute a normalization layer. More...
 
class  NENormalizationLayerKernel
 Interface for the normalization layer kernel. More...
 
class  NEOpticalFlow
 Basic function to execute optical flow. More...
 
class  NEPadLayer
 Basic function to pad a tensor. More...
 
class  NEPermute
 Basic function to run NEPermuteKernel. More...
 
class  NEPermuteKernel
 NEON kernel to perform tensor permutation. More...
 
class  NEPhase
 Basic function to run NEMagnitudePhaseKernel. More...
 
class  NEPixelWiseMultiplication
 Basic function to run NEPixelWiseMultiplicationKernel. More...
 
class  NEPixelWiseMultiplicationKernel
 Interface for the kernel to perform addition between two tensors. More...
 
class  NEPoolingLayer
 Basic function to simulate a pooling layer with the specified pooling operation. More...
 
class  NEPoolingLayerKernel
 Interface for the pooling layer kernel. More...
 
class  NEPowerOperationKernel
 
class  NEPReluLayer
 Basic function to run NEArithmeticOperationKernel for PRELU. More...
 
class  NEPriorBoxLayer
 Basic function to run NEPriorBoxLayerKernel. More...
 
class  NEPriorBoxLayerKernel
 Interface for the kernel to calculate prior boxes. More...
 
class  NEQuantizationLayer
 Basic function to simulate a quantization layer. More...
 
class  NEQuantizationLayerKernel
 Interface for the quantization layer kernel. More...
 
class  NERange
 Basic function to run NERangeKernel. More...
 
class  NERangeKernel
 Kernel class for Range. More...
 
class  NEReduceMean
 Basic function to perform reduce operation. More...
 
class  NEReductionOperation
 Basic function to simulate a reduction operation. More...
 
class  NEReductionOperationKernel
 NEON kernel to perform a reduction operation. More...
 
class  NERemap
 Basic function to execute remap. More...
 
class  NERemapKernel
 NEON kernel to perform a remap on a tensor. More...
 
class  NEReorgLayer
 Basic function to run NEReorgLayerKernel. More...
 
class  NEReorgLayerKernel
 Interface for the kernel to perform tensor re-organization. More...
 
class  NEReshapeLayer
 Basic function to run NEReshapeLayerKernel. More...
 
class  NEReshapeLayerKernel
 Interface for the kernel to perform tensor reshaping. More...
 
class  NEReverse
 Basic function to run NEReverseKernel. More...
 
class  NEReverseKernel
 Interface for the reverse layer kernel. More...
 
class  NERNNLayer
 Basic function to run NERNNLayer. More...
 
class  NEROIPoolingLayer
 Basic function to run NEROIPoolingLayerKernel. More...
 
class  NEROIPoolingLayerKernel
 Interface for the ROI pooling layer kernel. More...
 
class  NERoundLayer
 Basic function to compute the round value elementwise of an input tensor. More...
 
class  NERsqrtLayer
 Basic function to perform inverse square root on an input tensor. More...
 
class  NEScale
 Basic function to run NEScaleKernel. More...
 
class  NEScaleKernel
 NEON kernel to perform scaling on a tensor. More...
 
class  NEScharr3x3
 Basic function to execute scharr 3x3 filter. More...
 
class  NEScharr3x3Kernel
 Interface for the kernel to run a 3x3 Scharr filter on a tensor. More...
 
class  NESelect
 Basic function to run NESelect. More...
 
class  NESelectKernel
 Interface for the select kernel. More...
 
class  NESeparableConvolutionHorKernel
 Kernel for the Horizontal pass of a Separable Convolution. More...
 
class  NESeparableConvolutionVertKernel
 Kernel for the Vertical pass of a Separable Convolution. More...
 
class  NESimpleAssemblyFunction
 Basic interface for functions which have a single NEON GEMM wrapper kernel to run. More...
 
class  NESinLayer
 Basic function to compute the sine of an input tensor. More...
 
class  NESlice
 Basic function to perform tensor slicing. More...
 
class  NESobel3x3
 Basic function to execute sobel 3x3 filter. More...
 
class  NESobel3x3Kernel
 Interface for the kernel to run a 3x3 Sobel X filter on a tensor. More...
 
class  NESobel5x5
 Basic function to execute sobel 5x5 filter. More...
 
class  NESobel5x5HorKernel
 Interface for the kernel to run the horizontal pass of 5x5 Sobel filter on a tensor. More...
 
class  NESobel5x5VertKernel
 Interface for the kernel to run the vertical pass of 5x5 Sobel Y filter on a tensor. More...
 
class  NESobel7x7
 Basic function to execute sobel 7x7 filter. More...
 
class  NESobel7x7HorKernel
 Interface for the kernel to run the horizontal pass of 7x7 Sobel filter on a tensor. More...
 
class  NESobel7x7VertKernel
 Interface for the kernel to run the vertical pass of 7x7 Sobel Y filter on a tensor. More...
 
class  NESoftmaxLayer
 Basic function to compute a SoftmaxLayer. More...
 
class  NESpaceToBatchLayer
 Basic function to spatial divide a tensor. More...
 
class  NESpaceToBatchLayerKernel
 Interface for the space to batch kernel. More...
 
class  NESpaceToDepthLayer
 This function calls the following NEON kernels/functions: More...
 
class  NESpaceToDepthLayerKernel
 Interface for the space to depth kernel. More...
 
class  NESplit
 Basic function to split a tensor along a given axis. More...
 
class  NEStackLayer
 Basic function to stack tensors along an axis. More...
 
class  NEStackLayerKernel
 NEON kernel to stacks a rank-R tensor into one with rank-(R+1) along the axis dimension. More...
 
class  NEStridedSlice
 Basic function to run NEStridedSliceKernel. More...
 
class  NEStridedSliceKernel
 Interface for the kernel to perform tensor strided slicing. More...
 
class  NETableLookup
 Basic function to run NETableLookupKernel. More...
 
class  NETableLookupKernel
 Interface for the kernel to perform table lookup calculations. More...
 
class  NEThreshold
 Basic function to run NEThresholdKernel. More...
 
class  NEThresholdKernel
 Interface for the thresholding kernel. More...
 
class  NETile
 Basic function to run NETileKernel. More...
 
class  NETileKernel
 NEON kernel to perform a tile operation. More...
 
class  NETranspose
 Basic function to transpose a matrix on NEON. More...
 
class  NETransposeKernel
 NEON kernel which transposes the elements of a matrix. More...
 
class  NEUnstack
 Basic function to unpack a rank-R tensor into rank-(R-1) tensors. More...
 
class  NEUpsampleLayer
 Function to run upsample layer. More...
 
class  NEUpsampleLayerKernel
 Interface for the Upsample layer kernel. More...
 
class  NEWarpAffine
 Basic function to run NEWarpAffineKernel. More...
 
class  NEWarpAffineKernel
 Template interface for the kernel to compute warp affine. More...
 
class  NEWarpPerspective
 Basic function to run NEWarpPerspectiveKernel. More...
 
class  NEWarpPerspectiveKernel
 Template interface for the kernel to compute warp perspective. More...
 
class  NEWeightsReshapeKernel
 NEON kernel to perform reshaping on the weights used by convolution and locally connected layer. More...
 
class  NEWidthConcatenateLayerKernel
 Interface for the width concatenate kernel. More...
 
class  NEWinogradConvolutionLayer
 Basic function to simulate a convolution layer. More...
 
class  NEWinogradLayerConfiguration
 NEON kernel to perform Winograd. More...
 
class  NEWinogradLayerTransformInputKernel
 NEON kernel to perform Winograd input transform. More...
 
class  NEWinogradLayerTransformOutputKernel
 NEON kernel to perform Winograd output transform. More...
 
class  NEWinogradLayerTransformWeightsKernel
 NEON kernel to perform Winograd weights transform. More...
 
class  NEYOLOLayer
 Basic function to run NEYOLOLayerKernel. More...
 
class  NEYOLOLayerKernel
 Interface for the YOLO layer kernel. More...
 
class  NormalizationLayerInfo
 Normalization Layer Information class. More...
 
class  OffsetLifetimeManager
 Concrete class that tracks the lifetime of registered tensors and calculates the systems memory requirements in terms of a single blob and a list of offsets. More...
 
class  OffsetMemoryPool
 Offset based memory pool. More...
 
class  OMPScheduler
 Pool of threads to automatically split a kernel's execution among several threads. More...
 
struct  OpticalFlowParameters
 Parameters of Optical Flow algorithm. More...
 
class  PadStrideInfo
 Padding and stride information class. More...
 
class  PixelValue
 Class describing the value of a pixel for any image format. More...
 
class  PoolingLayerInfo
 Pooling Layer Information class. More...
 
class  PoolManager
 Memory pool manager. More...
 
class  PriorBoxLayerInfo
 PriorBox layer info. More...
 
class  Program
 Program class. More...
 
class  Pyramid
 Basic implementation of the pyramid interface. More...
 
class  PyramidInfo
 Store the Pyramid's metadata. More...
 
class  QuantizationInfo
 Quantization information. More...
 
struct  Rectangle
 Rectangle type. More...
 
class  ROIPoolingLayerInfo
 ROI Pooling Layer Information class. More...
 
class  Scheduler
 Configurable scheduler which supports multiple multithreading APIs and choosing between different schedulers at runtime. More...
 
class  Semaphore
 Semamphore class. More...
 
class  SingleThreadScheduler
 Pool of threads to automatically split a kernel's execution among several threads. More...
 
class  Size2D
 Class for specifying the size of an image or rectangle. More...
 
class  Status
 Status class. More...
 
class  Steps
 Class to describe a number of elements in each dimension. More...
 
class  Strides
 Strides of an item in bytes. More...
 
class  SubTensor
 Basic implementation of the sub-tensor interface. More...
 
class  SubTensorInfo
 Store the sub tensor's metadata. More...
 
class  Tensor
 Basic implementation of the tensor interface. More...
 
class  TensorAccessor
 Tensor accessors to make it easier to interface with arm_gemm. More...
 
class  TensorAllocator
 Basic implementation of a CPU memory tensor allocator. More...
 
class  TensorInfo
 Store the tensor's metadata. More...
 
class  TensorShape
 Shape of a tensor. More...
 
struct  ThreadInfo
 Information about executing thread and CPU. More...
 
struct  UniformQuantizationInfo
 Quantization info when assuming per layer quantization. More...
 
struct  ValidRegion
 Container for valid region of a window. More...
 
class  WeightsInfo
 Convolution Layer Weights Information class. More...
 
class  Window
 Describe a multidimensional execution window. More...
 
class  WindowIterator
 Iterate over a portion of a Window. More...
 
struct  WinogradInfo
 Winograd information. More...
 

Typedefs

using ICLKeyPointArray = ICLArray< KeyPoint >
 Interface for OpenCL Array of Key Points. More...
 
using ICLCoordinates2DArray = ICLArray< Coordinates2D >
 Interface for OpenCL Array of 2D Coordinates. More...
 
using ICLDetectionWindowArray = ICLArray< DetectionWindow >
 Interface for OpenCL Array of Detection Windows. More...
 
using ICLSize2DArray = ICLArray< Size2D >
 Interface for OpenCL Array of 2D Sizes. More...
 
using ICLUInt8Array = ICLArray< cl_uchar >
 Interface for OpenCL Array of uint8s. More...
 
using ICLUInt16Array = ICLArray< cl_ushort >
 Interface for OpenCL Array of uint16s. More...
 
using ICLUInt32Array = ICLArray< cl_uint >
 Interface for OpenCL Array of uint32s. More...
 
using ICLInt16Array = ICLArray< cl_short >
 Interface for OpenCL Array of int16s. More...
 
using ICLInt32Array = ICLArray< cl_int >
 Interface for OpenCL Array of int32s. More...
 
using ICLFloatArray = ICLArray< cl_float >
 Interface for OpenCL Array of floats. More...
 
using ICLImage = ICLTensor
 Interface for OpenCL images. More...
 
using CLConvolution3x3Kernel = CLConvolutionKernel< 3 >
 Interface for the kernel which applies a 3x3 convolution to a tensor. More...
 
using CLConvolution5x5Kernel = CLConvolutionKernel< 5 >
 Interface for the kernel which applies a 5x5 convolution to a tensor. More...
 
using CLConvolution7x7Kernel = CLConvolutionKernel< 7 >
 Interface for the kernel which applies a 7x7 convolution to a tensor. More...
 
using CLConvolution9x9Kernel = CLConvolutionKernel< 9 >
 Interface for the kernel which applies a 9x9 convolution to a tensor. More...
 
using CLSeparableConvolution5x5HorKernel = CLSeparableConvolutionHorKernel< 5 >
 Interface for the kernel which applies a horizontal pass of 5x5 convolution to a tensor. More...
 
using CLSeparableConvolution7x7HorKernel = CLSeparableConvolutionHorKernel< 7 >
 Interface for the kernel which applies a horizontal pass of 7x7 convolution to a tensor. More...
 
using CLSeparableConvolution9x9HorKernel = CLSeparableConvolutionHorKernel< 9 >
 Interface for the kernel which applies a horizontal pass of 9x9 convolution to a tensor. More...
 
using CLSeparableConvolution5x5VertKernel = CLSeparableConvolutionVertKernel< 5 >
 Interface for the kernel which applies a vertical pass of 5x5 convolution to a tensor. More...
 
using CLSeparableConvolution7x7VertKernel = CLSeparableConvolutionVertKernel< 7 >
 Interface for the kernel which applies a vertical pass of 7x7 convolution to a tensor. More...
 
using CLSeparableConvolution9x9VertKernel = CLSeparableConvolutionVertKernel< 9 >
 Interface for the kernel which applies a vertical pass of 9x9 convolution to a tensor. More...
 
using ICLLKInternalKeypointArray = ICLArray< CLLKInternalKeypoint >
 Interface for OpenCL Array of Internal Key Points. More...
 
using ICLCoefficientTableArray = ICLArray< CLCoefficientTable >
 Interface for OpenCL Array of Coefficient Tables. More...
 
using ICLOldValArray = ICLArray< CLOldValue >
 Interface for OpenCL Array of Old Values. More...
 
using IImage = ITensor
 Interface for CPP Images. More...
 
using IGCImage = IGCTensor
 Interface for GLES Compute image. More...
 
using GCDirectConvolutionLayer1x1Kernel = GCDirectConvolutionLayerKernel< 1 >
 Interface for the 1x1 direct convolution kernel. More...
 
using GCDirectConvolutionLayer3x3Kernel = GCDirectConvolutionLayerKernel< 3 >
 Interface for the 3x3 direct convolution kernel. More...
 
using GCDirectConvolutionLayer5x5Kernel = GCDirectConvolutionLayerKernel< 5 >
 Interface for the 5x5 direct convolution kernel. More...
 
using IKeyPointArray = IArray< KeyPoint >
 Interface for Array of Key Points. More...
 
using ICoordinates2DArray = IArray< Coordinates2D >
 Interface for Array of 2D Coordinates. More...
 
using IDetectionWindowArray = IArray< DetectionWindow >
 Interface for Array of Detection Windows. More...
 
using ISize2DArray = IArray< Size2D >
 Interface for Array of 2D Sizes. More...
 
using IUInt8Array = IArray< uint8_t >
 Interface for Array of uint8s. More...
 
using IUInt16Array = IArray< uint16_t >
 Interface for Array of uint16s. More...
 
using IUInt32Array = IArray< uint32_t >
 Interface for Array of uint32s. More...
 
using IInt16Array = IArray< int16_t >
 Interface for Array of int16s. More...
 
using IInt32Array = IArray< int32_t >
 Interface for Array of int32s. More...
 
using IFloatArray = IArray< float >
 Interface for Array of floats. More...
 
using INEKernel = ICPPKernel
 Common interface for all kernels implemented in NEON. More...
 
using INESimpleKernel = ICPPSimpleKernel
 Interface for simple NEON kernels having 1 tensor input and 1 tensor output. More...
 
using NEAccumulateWeightedFP16Kernel = NEAccumulateWeightedKernel
 Interface for the accumulate weighted kernel using F16. More...
 
using NEBox3x3FP16Kernel = NEBox3x3Kernel
 NEON kernel to perform a Box 3x3 filter for FP16 datatype. More...
 
using NEConvolution3x3Kernel = NEConvolutionKernel< 3 >
 Interface for the kernel which applied a 3x3 convolution to a tensor. More...
 
using NEConvolution5x5Kernel = NEConvolutionKernel< 5 >
 Interface for the kernel which applied a 5x5 convolution to a tensor. More...
 
using NEConvolution7x7Kernel = NEConvolutionKernel< 7 >
 Interface for the kernel which applied a 7x7 convolution to a tensor. More...
 
using NEConvolution9x9Kernel = NEConvolutionKernel< 9 >
 Interface for the kernel which applied a 9x9 convolution to a tensor. More...
 
using NESeparableConvolution5x5HorKernel = NESeparableConvolutionHorKernel< 5 >
 Interface for the kernel which applied a 5x1 horizontal convolution to a tensor. More...
 
using NESeparableConvolution7x7HorKernel = NESeparableConvolutionHorKernel< 7 >
 Interface for the kernel which applied a 7x1 horizontal convolution to a tensor. More...
 
using NESeparableConvolution9x9HorKernel = NESeparableConvolutionHorKernel< 9 >
 Interface for the kernel which applied a 9x1 horizontal convolution to a tensor. More...
 
using NESeparableConvolution5x5VertKernel = NESeparableConvolutionVertKernel< 5 >
 Interface for the kernel which applied a 1x5 vertical convolution to a tensor. More...
 
using NESeparableConvolution7x7VertKernel = NESeparableConvolutionVertKernel< 7 >
 Interface for the kernel which applied a 1x7 vertical convolution to a tensor. More...
 
using NESeparableConvolution9x9VertKernel = NESeparableConvolutionVertKernel< 9 >
 Interface for the kernel which applied a 1x9 vertical convolution to a tensor. More...
 
using INELKInternalKeypointArray = IArray< NELKInternalKeypoint >
 Interface for NEON Array of Internal Key Points. More...
 
using NENonMaximaSuppression3x3FP16Kernel = NENonMaximaSuppression3x3Kernel
 NEON kernel to perform Non-Maxima suppression 3x3 with intermediate results in FP16 if the input data type is FP32. More...
 
using qasymm8x8_t = uint8x8_t
 8 bit quantized asymmetric vector with 8 elements More...
 
using qasymm8x8x2_t = uint8x8x2_t
 8 bit quantized asymmetric vector with 16 elements More...
 
using qasymm8x8x3_t = uint8x8x3_t
 8 bit quantized asymmetric vector with 24 elements More...
 
using qasymm8x8x4_t = uint8x8x4_t
 8 bit quantized asymmetric vector with 32 elements More...
 
using qasymm8x16_t = uint8x16_t
 8 bit quantized asymmetric vector with 16 elements More...
 
using qsymm8_t = int8_t
 8 bit quantized symmetric scalar value More...
 
using qsymm16_t = int16_t
 16 bit quantized symmetric scalar value More...
 
using qsymm16x8_t = int16x8_t
 16 bit quantized symmetric vector with 8 elements More...
 
using qsymm16x8x2_t = int16x8x2_t
 16 bit quantized symmetric vector with 16 elements More...
 
using qasymm8_t = uint8_t
 8 bit quantized asymmetric scalar value More...
 
using half = half_float::half
 16-bit floating point type More...
 
using PermutationVector = Strides
 Permutation vector. More...
 
using BiStrides = Coordinates
 Bidirectional strides. More...
 
using PaddingSize = BorderSize
 Container for 2D padding size. More...
 
using InternalKeypoint = std::tuple< float, float, float >
 Internal key point. More...
 
using PaddingInfo = std::pair< uint32_t, uint32_t >
 Padding information as a pair of unsigned int start/end. More...
 
using PaddingList = std::vector< PaddingInfo >
 List of padding information. More...
 
using Multiples = std::vector< uint32_t >
 Information to produce a tiled version of a Tensor. More...
 
using BBox = std::array< float, 4 >
 
using LabelBBox = std::map< int, std::vector< BBox > >
 
using KeyPointArray = Array< KeyPoint >
 Array of Key Points. More...
 
using Coordinates2DArray = Array< Coordinates2D >
 Array of 2D Coordinates. More...
 
using DetectionWindowArray = Array< DetectionWindow >
 Array of Detection Windows. More...
 
using Size2DArray = Array< Size2D >
 Array of 2D Sizes. More...
 
using UInt8Array = Array< uint8_t >
 Array of uint8s. More...
 
using UInt16Array = Array< uint16_t >
 Array of uint16s. More...
 
using UInt32Array = Array< uint32_t >
 Array of uint32s. More...
 
using Int16Array = Array< int16_t >
 Array of int16s. More...
 
using Int32Array = Array< int32_t >
 Array of int32s. More...
 
using FloatArray = Array< float >
 Array of floats. More...
 
using CLKeyPointArray = CLArray< KeyPoint >
 OpenCL Array of Key Points. More...
 
using CLCoordinates2DArray = CLArray< Coordinates2D >
 OpenCL Array of 2D Coordinates. More...
 
using CLDetectionWindowArray = CLArray< DetectionWindow >
 OpenCL Array of Detection Windows. More...
 
using CLSize2DArray = CLArray< Size2D >
 OpenCL Array of 2D Sizes. More...
 
using CLUInt8Array = CLArray< cl_uchar >
 OpenCL Array of uint8s. More...
 
using CLUInt16Array = CLArray< cl_ushort >
 OpenCL Array of uint16s. More...
 
using CLUInt32Array = CLArray< cl_uint >
 OpenCL Array of uint32s. More...
 
using CLInt16Array = CLArray< cl_short >
 OpenCL Array of int16s. More...
 
using CLInt32Array = CLArray< cl_int >
 OpenCL Array of int32s. More...
 
using CLFloatArray = CLArray< cl_float >
 OpenCL Array of floats. More...
 
using CLMemoryGroup = MemoryGroupBase< CLTensor >
 Memory Group in OpenCL. More...
 
using CLImage = CLTensor
 OpenCL Image. More...
 
using CLEqual = CLComparisonStatic< ComparisonOperation::Equal >
 Basic function to run equal comparison. More...
 
using CLNotEqual = CLComparisonStatic< ComparisonOperation::NotEqual >
 Basic function to run not equal comparison. More...
 
using CLGreater = CLComparisonStatic< ComparisonOperation::Greater >
 Basic function to run greater comparison. More...
 
using CLGreaterEqual = CLComparisonStatic< ComparisonOperation::GreaterEqual >
 Basic function to run greater-equal comparison. More...
 
using CLLess = CLComparisonStatic< ComparisonOperation::Less >
 Basic function to run less comparison. More...
 
using CLLessEqual = CLComparisonStatic< ComparisonOperation::LessEqual >
 Basic function to run less-equal comparison. More...
 
using CLConvolution5x5 = CLConvolutionSquare< 5 >
 Basic function to run 5x5 convolution. More...
 
using CLConvolution7x7 = CLConvolutionSquare< 7 >
 Basic function to run 7x7 convolution. More...
 
using CLConvolution9x9 = CLConvolutionSquare< 9 >
 Basic function to run 9x9 convolution. More...
 
using CLLKInternalKeypointArray = CLArray< CLLKInternalKeypoint >
 OpenCL Array of Internal Keypoints. More...
 
using CLCoefficientTableArray = CLArray< CLCoefficientTable >
 OpenCL Array of Coefficient Tables. More...
 
using CLOldValueArray = CLArray< CLOldValue >
 OpenCL Array of Old Values. More...
 
using GCMemoryGroup = MemoryGroupBase< GCTensor >
 
using GCImage = GCTensor
 OpenGL ES Image. More...
 
using MemoryGroup = MemoryGroupBase< Tensor >
 Memory Group. More...
 
using NEConvolution5x5 = NEConvolutionSquare< 5 >
 Basic function to run 5x5 convolution. More...
 
using NEConvolution7x7 = NEConvolutionSquare< 7 >
 Basic function to run 7x7 convolution. More...
 
using NEConvolution9x9 = NEConvolutionSquare< 9 >
 Basic function to run 9x9 convolution. More...
 
using NEEqual = NEElementwiseComparisonStatic< ComparisonOperation::Equal >
 Basic function to run equal comparison. More...
 
using NENotEqual = NEElementwiseComparisonStatic< ComparisonOperation::NotEqual >
 Basic function to run not equal comparison. More...
 
using NEGreater = NEElementwiseComparisonStatic< ComparisonOperation::Greater >
 Basic function to run greater comparison. More...
 
using NEGreaterEqual = NEElementwiseComparisonStatic< ComparisonOperation::GreaterEqual >
 Basic function to run greater-equal comparison. More...
 
using NELess = NEElementwiseComparisonStatic< ComparisonOperation::Less >
 Basic function to run less comparison. More...
 
using NELessEqual = NEElementwiseComparisonStatic< ComparisonOperation::LessEqual >
 Basic function to run less-equal comparison. More...
 
using LKInternalKeypointArray = Array< NELKInternalKeypoint >
 Array of LK Internel Keypoints. More...
 
using NEScheduler = Scheduler
 NEON Scheduler. More...
 
using Image = Tensor
 Image. More...
 
using MemoryMappings = std::map< IMemory *, size_t >
 A map of (handle, index/offset), where handle is the memory handle of the object to provide the memory for and index/offset is the buffer/offset from the pool that should be used. More...
 
using GroupMappings = std::map< size_t, MemoryMappings >
 A map of the groups and memory mappings. More...
 
template<typename T , int N>
using vec_n_t = typename vec_n_type< T, N >::type
 
template<typename T , int N>
using vec_n_byte_t = vec_n_t< T, N/sizeof(T) >
 
template<typename T >
using vec_16_byte_t = vec_n_byte_t< T, 16 >
 
template<typename T >
using vec_8_byte_t = vec_n_byte_t< T, 8 >
 
template<typename T >
using const_ptr_t = const T *
 
template<typename T >
using ptr_t = T *
 
template<typename V >
using elem_type_t = decltype(vget_lane< 0 >(std::declval< V >()))
 
using Mutex = std::mutex
 Wrapper of Mutex data-object. More...
 

Enumerations

enum  CLVersion {
  CL10, CL11, CL12, CL20,
  UNKNOWN
}
 Available OpenCL Version. More...
 
enum  CPUModel {
  GENERIC, GENERIC_FP16, GENERIC_FP16_DOT, A53,
  A55r0, A55r1
}
 CPU models - we only need to detect CPUs we have microarchitecture-specific code for. More...
 
enum  MemoryPolicy { MINIMIZE, NORMAL }
 Global memory policy. More...
 
enum  ErrorCode { OK, RUNTIME_ERROR, UNSUPPORTED_EXTENSION_USE }
 Available error codes. More...
 
enum  GPUTarget {
  UNKNOWN = 0x101, GPU_ARCH_MASK = 0xF00, MIDGARD = 0x100, BIFROST = 0x200,
  VALHALL = 0x300, T600 = 0x110, T700 = 0x120, T800 = 0x130,
  G71 = 0x210, G72 = 0x220, G51 = 0x230, G51BIG = 0x231,
  G51LIT = 0x232, G52 = 0x240, G52LIT = 0x241, G76 = 0x250,
  G77 = 0x310, TBOX = 0x320, TODX = 0x330
}
 Available GPU Targets. More...
 
enum  RoundingPolicy { TO_ZERO, TO_NEAREST_UP, TO_NEAREST_EVEN }
 Rounding method. More...
 
enum  Format {
  UNKNOWN, U8, S16, U16,
  S32, U32, F16, F32,
  UV88, RGB888, RGBA8888, YUV444,
  YUYV422, NV12, NV21, IYUV,
  UYVY422
}
 Image colour formats. More...
 
enum  DataType {
  UNKNOWN, U8, S8, QSYMM8,
  QASYMM8, QSYMM8_PER_CHANNEL, U16, S16,
  QSYMM16, U32, S32, U64,
  S64, F16, F32, F64,
  SIZET
}
 Available data types. More...
 
enum  SamplingPolicy { CENTER, TOP_LEFT }
 Available Sampling Policies. More...
 
enum  DataLayout { UNKNOWN, NCHW, NHWC }
 [DataLayout enum definition] More...
 
enum  DataLayoutDimension { CHANNEL, HEIGHT, WIDTH, BATCHES }
 [DataLayout enum definition] More...
 
enum  ConvolutionMethod { GEMM, DIRECT, WINOGRAD, FFT }
 Available ConvolutionMethod. More...
 
enum  DeconvolutionMethod { GEMM, DIRECT }
 Available DeconvolutionMethod. More...
 
enum  FuseBatchNormalizationType { CONVOLUTION, DEPTHWISECONVOLUTION }
 Available FuseBatchNormalizationType. More...
 
enum  PaddingMode { CONSTANT, REFLECT, SYMMETRIC }
 Padding mode to use for PadLayer. More...
 
enum  ComparisonOperation {
  Equal, NotEqual, Greater, GreaterEqual,
  Less, LessEqual
}
 Supported comparison operations. More...
 
enum  BorderMode { UNDEFINED, CONSTANT, REPLICATE }
 Methods available to handle borders. More...
 
enum  ConvertPolicy { WRAP, SATURATE }
 Policy to handle overflow. More...
 
enum  InterpolationPolicy { NEAREST_NEIGHBOR, BILINEAR, AREA }
 Interpolation method. More...
 
enum  BilinearInterpolation { BILINEAR_OLD_NEW, BILINEAR_SCHARR }
 Bilinear Interpolation method used by LKTracker. More...
 
enum  ThresholdType { BINARY, RANGE }
 Threshold mode. More...
 
enum  Termination { TERM_CRITERIA_EPSILON, TERM_CRITERIA_ITERATIONS, TERM_CRITERIA_BOTH }
 Termination criteria. More...
 
enum  MagnitudeType { L1NORM, L2NORM }
 Magnitude calculation type. More...
 
enum  PhaseType { SIGNED, UNSIGNED }
 Phase calculation type. More...
 
enum  Channel {
  UNKNOWN, C0, C1, C2,
  C3, R, G, B,
  A, Y, U, V
}
 Available channels. More...
 
enum  MatrixPattern { BOX, CROSS, DISK, OTHER }
 Available matrix patterns. More...
 
enum  NonLinearFilterFunction : unsigned { MEDIAN = 0, MIN = 1, MAX = 2 }
 Available non linear functions. More...
 
enum  ReductionOperation {
  ARG_IDX_MAX, ARG_IDX_MIN, MEAN_SUM, PROD,
  SUM_SQUARE, SUM, MIN, MAX
}
 Available reduction operations. More...
 
enum  ArithmeticOperation {
  ADD, SUB, DIV, MIN,
  MAX, SQUARED_DIFF, POWER, PRELU
}
 Available element-wise operations. More...
 
enum  ElementWiseUnary {
  RSQRT, EXP, NEG, LOG,
  ABS, SIN, ROUND
}
 Available element wise unary operations. More...
 
enum  NormType { IN_MAP_1D, IN_MAP_2D, CROSS_MAP }
 The normalization type used for the normalization layer. More...
 
enum  HOGNormType { L2_NORM = 1, L2HYS_NORM = 2, L1_NORM = 3 }
 Normalization type for Histogram of Oriented Gradients (HOG) More...
 
enum  DimensionRoundingType { FLOOR, CEIL }
 Dimension rounding type when down-scaling on CNNs. More...
 
enum  PoolingType { MAX, AVG, L2 }
 Available pooling types. More...
 
enum  NMSType { LINEAR, GAUSSIAN, ORIGINAL }
 Available non maxima suppression types. More...
 
enum  DetectionOutputLayerCodeType { CORNER, CENTER_SIZE, CORNER_SIZE, TF_CENTER }
 Available Detection Output code types. More...
 
enum  GEMMLowpOutputStageType { NONE, QUANTIZE_DOWN, QUANTIZE_DOWN_FIXEDPOINT, QUANTIZE_DOWN_FLOAT }
 GEMMLowp output stage type. More...
 
enum  CLTunerMode { EXHAUSTIVE, NORMAL, RAPID }
 < OpenCL tuner modes More...
 
enum  FFTDirection { Forward, Inverse }
 FFT direction to use. More...
 
enum  MappingType { BLOBS, OFFSETS }
 Mapping type. More...
 
enum  GradientDimension { GRAD_XY }
 Gradient dimension type. More...
 

Functions

std::string get_cl_type_from_data_type (const DataType &dt)
 Translates a tensor data type to the appropriate OpenCL type. More...
 
std::string get_cl_select_type_from_data_type (const DataType &dt)
 Translates a tensor data type to the appropriate OpenCL select type. More...
 
std::string get_data_size_from_data_type (const DataType &dt)
 Get the size of a data type in number of bits. More...
 
std::string get_underlying_cl_type_from_data_type (const DataType &dt)
 Translates fixed point tensor data type to the underlying OpenCL type. More...
 
GPUTarget get_target_from_device (const cl::Device &device)
 Helper function to get the GPU target from CL device. More...
 
CLVersion get_cl_version (const cl::Device &device)
 Helper function to get the highest OpenCL version supported. More...
 
bool device_supports_extension (const cl::Device &device, const char *extension_name)
 Helper function to check whether a given extension is supported. More...
 
bool fp16_supported (const cl::Device &device)
 Helper function to check whether the cl_khr_fp16 extension is supported. More...
 
bool arm_non_uniform_workgroup_supported (const cl::Device &device)
 Helper function to check whether the arm_non_uniform_work_group_size extension is supported. More...
 
bool dot8_supported (const cl::Device &device)
 Helper function to check whether the cl_arm_integer_dot_product_int8 extension is supported. More...
 
bool dot8_acc_supported (const cl::Device &device)
 Helper function to check whether the cl_arm_integer_dot_product_accumulate_int8 extension is supported. More...
 
bool cl_winograd_convolution_layer_supported (const Size2D &output_tile, const Size2D &kernel_size, DataLayout data_layout)
 This function checks if the Winograd configuration (defined through the output tile, kernel size and the data layout) is supported on OpenCL. More...
 
size_t preferred_vector_width (const cl::Device &device, DataType dt)
 Helper function to get the preferred native vector width size for built-in scalar types that can be put into vectors. More...
 
bool preferred_dummy_work_items_support (const cl::Device &device)
 Helper function to check if "dummy work-items" are preferred to have a power of two NDRange In case dummy work-items is enabled, it is OpenCL kernel responsibility to check if the work-item is out-of range or not. More...
 
arm_compute::Status error_on_unsupported_int64_base_atomics (const char *function, const char *file, const int line)
 Return an error if int64_base_atomics extension is not supported by the device. More...
 
void enqueue (cl::CommandQueue &queue, ICLKernel &kernel, const Window &window, const cl::NDRange &lws_hint=CLKernelLibrary::get().default_ndrange(), bool use_dummy_work_items=false)
 Add the kernel to the command queue with the given window. More...
 
bool opencl_is_available ()
 Check if OpenCL is available. More...
 
std::string cpu_model_to_string (CPUModel val)
 Convert a cpumodel value to a string. More...
 
arm_compute::Status error_on_unsupported_cpu_fp16 (const char *function, const char *file, const int line, const ITensorInfo *tensor_info)
 Return an error if the data type of the passed tensor info is FP16 and FP16 support is not compiled in. More...
 
arm_compute::Status error_on_unsupported_cpu_fp16 (const char *function, const char *file, const int line, const ITensor *tensor)
 Return an error if the data type of the passed tensor is FP16 and FP16 support is not compiled in. More...
 
template<typename T >
bool operator== (const Dimensions< T > &lhs, const Dimensions< T > &rhs)
 Check that given dimensions are equal. More...
 
template<typename T >
bool operator!= (const Dimensions< T > &lhs, const Dimensions< T > &rhs)
 Check that given dimensions are not equal. More...
 
template<typename... T>
void ignore_unused (T &&...)
 Ignores unused arguments. More...
 
Status create_error_va_list (ErrorCode error_code, const char *function, const char *file, const int line, const char *msg, va_list args)
 Creates an error containing the error message from variable argument list. More...
 
Status create_error (ErrorCode error_code, const char *function, const char *file, const int line, const char *msg,...)
 Creates an error containing the error message. More...
 
void error (const char *function, const char *file, const int line, const char *msg,...)
 Print an error message then throw an std::runtime_error. More...
 
GPUTarget get_target_from_device ()
 Helper function to get the GPU target from GLES using GL_RENDERER enum. More...
 
void enqueue (IGCKernel &kernel, const Window &window, const gles::NDRange &lws=gles::NDRange(1U, 1U, 1U))
 Add the kernel to the command queue with the given window. More...
 
bool opengles31_is_available ()
 Check if the OpenGL ES 3.1 API is available at runtime. More...
 
const std::string & string_from_target (GPUTarget target)
 Translates a given gpu device target to string. More...
 
GPUTarget get_target_from_name (const std::string &device_name)
 Helper function to get the GPU target from a device name. More...
 
GPUTarget get_arch_from_target (GPUTarget target)
 Helper function to get the GPU arch. More...
 
template<typename... Args>
bool gpu_target_is_in (GPUTarget target_to_check, GPUTarget target, Args... targets)
 Helper function to check whether a gpu target is equal to the provided targets. More...
 
bool gpu_target_is_in (GPUTarget target_to_check, GPUTarget target)
 Variant of gpu_target_is_in for comparing two targets. More...
 
template<typename Kernel , typename... T>
std::unique_ptr< Kernelcreate_configure_kernel (T &&... args)
 Helper function to create and return a unique_ptr pointed to a CL/GLES kernel object It also calls the kernel's configuration. More...
 
template<typename Kernel >
std::unique_ptr< Kernelcreate_kernel ()
 Helper function to create and return a unique_ptr pointed to a CL/GLES kernel object. More...
 
template<typename T >
delta_bilinear_c1 (const T *pixel_ptr, size_t stride, float dx, float dy)
 Computes bilinear interpolation using the pointer to the top-left pixel and the pixel's distance between the real coordinates and the smallest following integer coordinates. More...
 
uint8_t delta_bilinear_c1_quantized (const uint8_t *pixel_ptr, size_t stride, float dx, float dy, UniformQuantizationInfo iq_info, UniformQuantizationInfo oq_info)
 Computes bilinear interpolation for quantized input and output, using the pointer to the top-left pixel and the pixel's distance between the real coordinates and the smallest following integer coordinates. More...
 
template<typename T >
delta_linear_c1_y (const T *pixel_ptr, size_t stride, float dy)
 Computes linear interpolation using the pointer to the top pixel and the pixel's distance between the real coordinates and the smallest following integer coordinates. More...
 
template<typename T >
delta_linear_c1_x (const T *pixel_ptr, float dx)
 Computes linear interpolation using the pointer to the left pixel and the pixel's distance between the real coordinates and the smallest following integer coordinates. More...
 
template<typename T >
pixel_bilinear_c1 (const T *first_pixel_ptr, size_t stride, float x, float y)
 Return the pixel at (x,y) using bilinear interpolation. More...
 
template<typename T >
uint8_t pixel_bilinear_c1_clamp (const T *first_pixel_ptr, size_t stride, size_t width, size_t height, float x, float y)
 Return the pixel at (x,y) using bilinear interpolation by clamping when out of borders. More...
 
uint8_t pixel_area_c1u8_clamp (const uint8_t *first_pixel_ptr, size_t stride, size_t width, size_t height, float wr, float hr, int x, int y)
 Return the pixel at (x,y) using area interpolation by clamping when out of borders. More...
 
template<typename L , typename... Ts>
void execute_window_loop (const Window &w, L &&lambda_function, Ts &&... iterators)
 Iterate through the passed window, automatically adjusting the iterators and calling the lambda_functino for each element. More...
 
template<typename... Ts>
bool update_window_and_padding (Window &win, Ts &&... patterns)
 Update window and padding size for each of the access patterns. More...
 
Window calculate_max_window (const ValidRegion &valid_region, const Steps &steps=Steps(), bool skip_border=false, BorderSize border_size=BorderSize())
 Calculate the maximum window for a given tensor shape and border setting. More...
 
Window calculate_max_window (const ITensorInfo &info, const Steps &steps=Steps(), bool skip_border=false, BorderSize border_size=BorderSize())
 Calculate the maximum window for a given tensor shape and border setting. More...
 
Window calculate_max_window_horizontal (const ValidRegion &valid_region, const Steps &steps=Steps(), bool skip_border=false, BorderSize border_size=BorderSize())
 Calculate the maximum window used by a horizontal kernel for a given tensor shape and border setting. More...
 
Window calculate_max_window_horizontal (const ITensorInfo &info, const Steps &steps=Steps(), bool skip_border=false, BorderSize border_size=BorderSize())
 Calculate the maximum window used by a horizontal kernel for a given tensor shape and border setting. More...
 
Window calculate_max_enlarged_window (const ValidRegion &valid_region, const Steps &steps=Steps(), BorderSize border_size=BorderSize())
 Calculate the maximum window for a given tensor shape and border setting. More...
 
Window calculate_max_enlarged_window (const ITensorInfo &info, const Steps &steps=Steps(), BorderSize border_size=BorderSize())
 Calculate the maximum window for a given tensor shape and border setting. More...
 
template<typename... Ts>
ValidRegion intersect_valid_regions (const Ts &... regions)
 Intersect multiple valid regions. More...
 
template<typename T , typename... Ts>
Strides compute_strides (const ITensorInfo &info, T stride_x, Ts &&... fixed_strides)
 Create a strides object based on the provided strides and the tensor dimensions. More...
 
template<typename... Ts>
Strides compute_strides (const ITensorInfo &info)
 Create a strides object based on the tensor dimensions. More...
 
template<typename T >
void permute (Dimensions< T > &dimensions, const PermutationVector &perm)
 Permutes given Dimensions according to a permutation vector. More...
 
void permute (TensorShape &shape, const PermutationVector &perm)
 Permutes given TensorShape according to a permutation vector. More...
 
bool auto_init_if_empty (ITensorInfo &info, const TensorShape &shape, int num_channels, DataType data_type, QuantizationInfo quantization_info=QuantizationInfo())
 Auto initialize the tensor info (shape, number of channels and data type) if the current assignment is empty. More...
 
bool auto_init_if_empty (ITensorInfo &info_sink, const ITensorInfo &info_source)
 Auto initialize the tensor info using another tensor info. More...
 
bool set_shape_if_empty (ITensorInfo &info, const TensorShape &shape)
 Set the shape to the specified value if the current assignment is empty. More...
 
bool set_format_if_unknown (ITensorInfo &info, Format format)
 Set the format, data type and number of channels to the specified value if the current data type is unknown. More...
 
bool set_data_type_if_unknown (ITensorInfo &info, DataType data_type)
 Set the data type and number of channels to the specified value if the current data type is unknown. More...
 
bool set_data_layout_if_unknown (ITensorInfo &info, DataLayout data_layout)
 Set the data layout to the specified value if the current data layout is unknown. More...
 
bool set_quantization_info_if_empty (ITensorInfo &info, QuantizationInfo quantization_info)
 Set the quantization info to the specified value if the current quantization info is empty and the data type of asymmetric quantized type. More...
 
ValidRegion calculate_valid_region_scale (const ITensorInfo &src_info, const TensorShape &dst_shape, InterpolationPolicy interpolate_policy, SamplingPolicy sampling_policy, bool border_undefined)
 Helper function to calculate the Valid Region for Scale. More...
 
Coordinates index2coords (const TensorShape &shape, int index)
 Convert a linear index into n-dimensional coordinates. More...
 
int coords2index (const TensorShape &shape, const Coordinates &coord)
 Convert n-dimensional coordinates into a linear index. More...
 
size_t get_data_layout_dimension_index (const DataLayout data_layout, const DataLayoutDimension data_layout_dimension)
 Get the index of the given dimension. More...
 
DataLayoutDimension get_index_data_layout_dimension (const DataLayout data_layout, const size_t index)
 Get the DataLayoutDimension of a given index and layout. More...
 
unsigned int get_normalization_dimension_index (DataLayout layout, const NormalizationLayerInfo &info)
 Calculate the normalization dimension index for a given normalization type. More...
 
Size2D compute_winograd_convolution_tiles (const Size2D &in_dims, const Size2D &kernel_size, const Size2D &output_tile_size, const PadStrideInfo &conv_info)
 Calculate the number of output tiles required by Winograd Convolution layer. More...
 
template<typename T >
wrap_around (T x, T m)
 Wrap-around a number within the range 0 <= x < m. More...
 
unsigned int get_next_power_two (unsigned int x)
 Given an integer value, this function returns the next power of two. More...
 
int adjust_down (int required, int available, int step)
 Decrease required in steps of step until it's less than available. More...
 
int adjust_up (int required, int available, int step)
 Increase required in steps of step until it's greater than available. More...
 
uint8x16_t vmlaq_qasymm8 (qasymm8x16_t vd, float32x4_t vs, float32x4_t vo)
 Perform a multiply-accumulate on all 16 components of a QASYMM8 vector. More...
 
template<bool is_bounded_relu>
uint8x16_t finalize_quantization (int32x4x4_t &in_s32, int result_fixedpoint_multiplier, int32_t result_shift, int32x4_t result_offset_after_shift_s32, uint8x16_t min_u8, uint8x16_t max_u8)
 Performs final quantization step on 16 elements. More...
 
template<bool is_bounded_relu>
uint8_t finalize_quantization (int32_t in_value, int result_fixedpoint_multiplier, int32_t result_shift, int32_t result_offset_after_shift_s32, uint8_t min_u8, uint8_t max_u8)
 Performs final quantization step on single element. More...
 
float32x4x2_t vdequantize (const uint8x8_t &qv, const UniformQuantizationInfo &qi)
 Dequantize a neon vector holding 8 quantized values. More...
 
float32x4x4_t vdequantize (const uint8x16_t &qv, const UniformQuantizationInfo &qi)
 Dequantize a neon vector holding 16 quantized values. More...
 
float32x4x4_t vdequantize (const uint8x16_t &qv, float scale, int32_t offset)
 Dequantize following an asymmetric quantization scheme a neon vector holding 16 quantized values. More...
 
float32x4x4_t vdequantize (const int8x16_t &qv, float scale)
 Dequantize following a symmetric quantization scheme a neon vector holding 16 quantized values. More...
 
uint8x8_t vquantize (const float32x4x2_t &qv, const UniformQuantizationInfo &qi)
 Quantize a neon vector holding 8 floating point values. More...
 
uint8x16_t vquantize (const float32x4x4_t &qv, const UniformQuantizationInfo &qi)
 Quantize a neon vector holding 16 floating point values. More...
 
void colorconvert_rgb_to_rgbx (const void *__restrict input, void *__restrict output, const Window &win)
 Convert RGB to RGBX. More...
 
void colorconvert_rgb_to_u8 (const void *__restrict input, void *__restrict output, const Window &win)
 Convert RGB to U8. More...
 
void colorconvert_rgbx_to_rgb (const void *input, void *output, const Window &win)
 Convert RGBX to RGB. More...
 
template<bool yuyv, bool alpha>
void colorconvert_yuyv_to_rgb (const void *__restrict input, void *__restrict output, const Window &win)
 Convert YUYV to RGB. More...
 
template<bool uv, bool alpha>
void colorconvert_nv12_to_rgb (const void *__restrict input, void *__restrict output, const Window &win)
 Convert NV12 to RGB. More...
 
template<bool alpha>
void colorconvert_iyuv_to_rgb (const void *__restrict input, void *__restrict output, const Window &win)
 Convert IYUV to RGB. More...
 
template<bool yuyv>
void colorconvert_yuyv_to_nv12 (const void *__restrict input, void *__restrict output, const Window &win)
 Convert YUYV to NV12. More...
 
void colorconvert_iyuv_to_nv12 (const void *__restrict input, void *__restrict output, const Window &win)
 Convert IYUV to NV12. More...
 
template<bool uv>
void colorconvert_nv12_to_iyuv (const void *__restrict input, void *__restrict output, const Window &win)
 Convert NV12 to IYUV. More...
 
template<bool yuyv>
void colorconvert_yuyv_to_iyuv (const void *__restrict input, void *__restrict output, const Window &win)
 Convert YUYV to IYUV. More...
 
template<bool uv>
void colorconvert_nv12_to_yuv4 (const void *__restrict input, void *__restrict output, const Window &win)
 Convert NV12 to YUV4. More...
 
void colorconvert_iyuv_to_yuv4 (const void *__restrict input, void *__restrict output, const Window &win)
 Convert IYUV to YUV4. More...
 
template<bool alpha>
void colorconvert_rgb_to_nv12 (const void *__restrict input, void *__restrict output, const Window &win)
 Convert RGB to NV12. More...
 
template<bool alpha>
void colorconvert_rgb_to_iyuv (const void *__restrict input, void *__restrict output, const Window &win)
 Convert RGB to IYUV. More...
 
template<bool alpha>
void colorconvert_rgb_to_yuv4 (const void *__restrict input, void *__restrict output, const Window &win)
 Convert RGB to YUV4. More...
 
float32x4x2_t vmax2q_f32 (float32x4x2_t a, float32x4x2_t b)
 Compute lane-by-lane maximum between elements of a float vector with 4x2 elements. More...
 
float32x4_t vfloorq_f32 (float32x4_t val)
 Calculate floor of a vector. More...
 
float32x4_t vroundq_rte_f32 (float32x4_t val)
 Calculate round value of a vector to nearest with ties to even. More...
 
float32x2_t vinvsqrt_f32 (float32x2_t x)
 Calculate inverse square root. More...
 
float32x4_t vinvsqrtq_f32 (float32x4_t x)
 Calculate inverse square root. More...
 
float32x2_t vinv_f32 (float32x2_t x)
 Calculate reciprocal. More...
 
float32x4_t vinvq_f32 (float32x4_t x)
 Calculate reciprocal. More...
 
float32x4_t vtaylor_polyq_f32 (float32x4_t x, const std::array< float32x4_t, 8 > &coeffs)
 Perform a 7th degree polynomial approximation using Estrin's method. More...
 
float32x4_t vexpq_f32 (float32x4_t x)
 Calculate exponential. More...
 
float32x4_t vlogq_f32 (float32x4_t x)
 Calculate logarithm. More...
 
float32x4_t vtanhq_f32 (float32x4_t val)
 Calculate hyperbolic tangent. More...
 
float32x4_t vpowq_f32 (float32x4_t val, float32x4_t n)
 Calculate n power of a number. More...
 
int32x4_t rounding_divide_by_pow2 (int32x4_t x, int exponent)
 Round to the nearest division by a power-of-two using exponent. More...
 
int32_t rounding_divide_by_pow2 (int32_t x, int exponent)
 Round to the nearest division by a power-of-two using exponent. More...
 
float32x4_t vsinq_f32 (float32x4_t val)
 Calculate sine. More...
 
float32x2_t vsin_f32 (float32x2_t val)
 Calculate sine. More...
 
template<bool is_bounded_relu>
int16x8_t finalize_quantization_int16 (int32x4x2_t &in_s32, int result_fixedpoint_multiplier, int32_t result_shift, int16x8_t min_s16, int16x8_t max_s16)
 Performs final quantization step on 8 signed 16-bit elements. More...
 
template<bool is_bounded_relu>
int16_t finalize_quantization_int16 (int32_t in_value, int result_fixedpoint_multiplier, int32_t result_shift, int16_t min_s16, int16_t max_s16)
 Performs final quantization step on single signed 16-bit element. More...
 
float32x4x2_t vdequantize_int16 (const int16x8_t &qv, float scale)
 Dequantize a neon vector holding 8 16-bit quantized values. More...
 
int16x8_t vquantize_int16 (const float32x4x2_t &qv, float scale)
 Quantize a neon vector holding 8 floating point values. More...
 
float32x4x4_t vdequantize (const int16x8x2_t &qv, const UniformQuantizationInfo &qi)
 Dequantize a neon vector holding 16 16-bit quantized values. More...
 
qsymm16x8x2_t vquantize_qsymm16 (const float32x4x4_t &qv, const UniformQuantizationInfo &qi)
 Quantize a neon vector holding 16 floating point values. More...
 
bool operator== (const QuantizationInfo &lhs, const QuantizationInfo &rhs)
 Check whether two quantization info are equal. More...
 
bool operator!= (const QuantizationInfo &lhs, const QuantizationInfo &rhs)
 Check whether two quantization info are not equal. More...
 
bool operator== (const UniformQuantizationInfo &lhs, const UniformQuantizationInfo &rhs)
 Check whether two quantization info are equal. More...
 
bool operator!= (const UniformQuantizationInfo &lhs, const UniformQuantizationInfo &rhs)
 Check whether two quantization info are not equal. More...
 
uint8_t quantize_qasymm8 (float value, const UniformQuantizationInfo &qinfo, RoundingPolicy rounding_policy=RoundingPolicy::TO_NEAREST_UP)
 Quantize a value given a asymmetric quantization scheme. More...
 
uint8_t quantize_qasymm8 (float value, const QuantizationInfo &qinfo, RoundingPolicy rounding_policy=RoundingPolicy::TO_NEAREST_UP)
 Quantize a value given a asymmetric quantization scheme. More...
 
int8_t quantize_qsymm8 (float value, const QuantizationInfo &qinfo)
 Quantize a value given a symmetric quantization scheme. More...
 
float dequantize_qasymm8 (uint8_t value, const UniformQuantizationInfo &qinfo)
 Dequantize a value given a asymmetric quantization scheme. More...
 
float dequantize_qasymm8 (uint8_t value, const QuantizationInfo &qinfo)
 Dequantize a value given a asymmetric quantization scheme. More...
 
float dequantize (uint8_t value, float scale, int32_t offset)
 Dequantize a value given an asymmetric quantization scheme. More...
 
float dequantize_qsymm8 (int8_t value, const UniformQuantizationInfo &qinfo)
 Dequantize a value given a symmetric quantization scheme. More...
 
float dequantize (int8_t value, float scale)
 Dequantize a value given a symmetric quantization scheme. More...
 
float dequantize (int16_t value, float scale)
 Dequantize a value given a symmetric quantization scheme. More...
 
int16_t quantize_qsymm16 (float value, const UniformQuantizationInfo &qinfo, RoundingPolicy rounding_policy=RoundingPolicy::TO_NEAREST_UP)
 Quantize a value given a 16-bit symmetric quantization scheme. More...
 
float dequantize_qsymm16 (int16_t value, const UniformQuantizationInfo &qinfo)
 Dequantize a value given a 16-bit symmetric quantization scheme. More...
 
int16_t quantize_qsymm16 (float value, const QuantizationInfo &qinfo)
 Quantize a value given a 16-bit symmetric quantization scheme. More...
 
float dequantize_qsymm16 (int16_t value, const QuantizationInfo &qinfo)
 Dequantize a value given a 16-bit symmetric quantization scheme. More...
 
int round (float x, RoundingPolicy rounding_policy)
 Return a rounded value of x. More...
 
template<typename S , typename T >
constexpr auto DIV_CEIL (S val, T m) -> decltype((val+m - 1)/m)
 Calculate the rounded up quotient of val / m. More...
 
template<typename S , typename T >
auto ceil_to_multiple (S value, T divisor) -> decltype(((value+divisor - 1)/divisor) *divisor)
 Computes the smallest number larger or equal to value that is a multiple of divisor. More...
 
template<typename S , typename T >
auto floor_to_multiple (S value, T divisor) -> decltype((value/divisor) *divisor)
 Computes the largest number smaller or equal to value that is a multiple of divisor. More...
 
std::string build_information ()
 Returns the arm_compute library build information. More...
 
std::string read_file (const std::string &filename, bool binary)
 Load an entire file in memory. More...
 
size_t data_size_from_type (DataType data_type)
 The size in bytes of the data type. More...
 
size_t pixel_size_from_format (Format format)
 The size in bytes of the pixel format. More...
 
size_t element_size_from_data_type (DataType dt)
 The size in bytes of the data type. More...
 
DataType data_type_from_format (Format format)
 Return the data type used by a given single-planar pixel format. More...
 
int plane_idx_from_channel (Format format, Channel channel)
 Return the plane index of a given channel given an input format. More...
 
int channel_idx_from_format (Format format, Channel channel)
 Return the channel index of a given channel given an input format. More...
 
size_t num_planes_from_format (Format format)
 Return the number of planes for a given format. More...
 
size_t num_channels_from_format (Format format)
 Return the number of channels for a given single-planar pixel format. More...
 
DataType get_promoted_data_type (DataType dt)
 Return the promoted data type of a given data type. More...
 
bool has_format_horizontal_subsampling (Format format)
 Return true if the given format has horizontal subsampling. More...
 
bool has_format_vertical_subsampling (Format format)
 Return true if the given format has vertical subsampling. More...
 
bool separate_matrix (const int16_t *conv, int16_t *conv_col, int16_t *conv_row, uint8_t size)
 Separate a 2D convolution into two 1D convolutions. More...
 
uint32_t calculate_matrix_scale (const int16_t *matrix, unsigned int matrix_size)
 Calculate the scale of the given square matrix. More...
 
TensorShape adjust_odd_shape (const TensorShape &shape, Format format)
 Adjust tensor shape size if width or height are odd for a given multi-planar format. More...
 
TensorShape calculate_subsampled_shape (const TensorShape &shape, Format format, Channel channel=Channel::UNKNOWN)
 Calculate subsampled shape for a given format and channel. More...
 
std::pair< DataType, DataTypedata_type_for_convolution (const int16_t *conv_col, const int16_t *conv_row, size_t size)
 Calculate accurary required by the horizontal and vertical convolution computations. More...
 
DataType data_type_for_convolution_matrix (const int16_t *conv, size_t size)
 Calculate the accuracy required by the squared convolution calculation. More...
 
template<typename T >
void permute_strides (Dimensions< T > &dimensions, const PermutationVector &perm)
 Permutes the given dimensions according the permutation vector. More...
 
PadStrideInfo calculate_same_pad (TensorShape input_shape, TensorShape weights_shape, PadStrideInfo conv_info, DataLayout data_layout=DataLayout::NCHW, const Size2D &dilation=Size2D(1u, 1u), const DimensionRoundingType &rounding_type=DimensionRoundingType::FLOOR)
 Calculate padding requirements in case of SAME padding. More...
 
std::pair< unsigned int, unsigned int > deconvolution_output_dimensions (unsigned int in_width, unsigned int in_height, unsigned int kernel_width, unsigned int kernel_height, unsigned int padx, unsigned int pady, unsigned int stride_x, unsigned int stride_y)
 Returns expected width and height of the deconvolution's output tensor. More...
 
std::pair< unsigned int, unsigned int > scaled_dimensions (unsigned int width, unsigned int height, unsigned int kernel_width, unsigned int kernel_height, const PadStrideInfo &pad_stride_info, const Size2D &dilation=Size2D(1U, 1U))
 Returns expected width and height of output scaled tensor depending on dimensions rounding mode. More...
 
const std::string & string_from_format (Format format)
 Convert a tensor format into a string. More...
 
const std::string & string_from_channel (Channel channel)
 Convert a channel identity into a string. More...
 
const std::string & string_from_data_layout (DataLayout dl)
 Convert a data layout identity into a string. More...
 
const std::string & string_from_data_type (DataType dt)
 Convert a data type identity into a string. More...
 
const std::string & string_from_matrix_pattern (MatrixPattern pattern)
 Convert a matrix pattern into a string. More...
 
const std::string & string_from_activation_func (ActivationLayerInfo::ActivationFunction act)
 Translates a given activation function to a string. More...
 
const std::string & string_from_non_linear_filter_function (NonLinearFilterFunction function)
 Translates a given non linear function to a string. More...
 
const std::string & string_from_interpolation_policy (InterpolationPolicy policy)
 Translates a given interpolation policy to a string. More...
 
const std::string & string_from_border_mode (BorderMode border_mode)
 Translates a given border mode policy to a string. More...
 
const std::string & string_from_norm_type (NormType type)
 Translates a given normalization type to a string. More...
 
const std::string & string_from_pooling_type (PoolingType type)
 Translates a given pooling type to a string. More...
 
const std::string & string_from_gemmlowp_output_stage (GEMMLowpOutputStageType output_stage)
 Translates a given GEMMLowp output stage to a string. More...
 
std::string string_from_pixel_value (const PixelValue &value, const DataType data_type)
 Convert a PixelValue to a string, represented through the specific data type. More...
 
std::string lower_string (const std::string &val)
 Lower a given string. More...
 
bool is_data_type_float (DataType dt)
 Check if a given data type is of floating point type. More...
 
bool is_data_type_quantized (DataType dt)
 Check if a given data type is of quantized type. More...
 
bool is_data_type_quantized_asymmetric (DataType dt)
 Check if a given data type is of asymmetric quantized type. More...
 
bool is_data_type_quantized_symmetric (DataType dt)
 Check if a given data type is of symmetric quantized type. More...
 
std::string float_to_string_with_full_precision (float val)
 Create a string with the float in full precision. More...
 
size_t num_of_elements_in_range (const float start, const float end, const float step)
 Returns the number of elements required to go from start to end with the wanted step. More...
 
template<typename T >
bool check_value_range (T val, DataType dt, QuantizationInfo qinfo=QuantizationInfo())
 Returns true if the value can be represented by the given data type. More...
 
template<typename... Ts>
arm_compute::Status error_on_nullptr (const char *function, const char *file, const int line, Ts &&... pointers)
 Create an error if one of the pointers is a nullptr. More...
 
arm_compute::Status error_on_mismatching_windows (const char *function, const char *file, const int line, const Window &full, const Window &win)
 Return an error if the passed window is invalid. More...
 
arm_compute::Status error_on_invalid_subwindow (const char *function, const char *file, const int line, const Window &full, const Window &sub)
 Return an error if the passed subwindow is invalid. More...
 
arm_compute::Status error_on_window_not_collapsable_at_dimension (const char *function, const char *file, const int line, const Window &full, const Window &window, const int dim)
 Return an error if the window can't be collapsed at the given dimension. More...
 
arm_compute::Status error_on_coordinates_dimensions_gte (const char *function, const char *file, const int line, const Coordinates &pos, unsigned int max_dim)
 Return an error if the passed coordinates have too many dimensions. More...
 
arm_compute::Status error_on_window_dimensions_gte (const char *function, const char *file, const int line, const Window &win, unsigned int max_dim)
 Return an error if the passed window has too many dimensions. More...
 
template<typename T , typename... Ts>
arm_compute::Status error_on_mismatching_dimensions (const char *function, const char *file, int line, const Dimensions< T > &dim1, const Dimensions< T > &dim2, Ts &&... dims)
 Return an error if the passed dimension objects differ. More...
 
template<typename... Ts>
arm_compute::Status error_on_tensors_not_even (const char *function, const char *file, int line, const Format &format, const ITensor *tensor1, Ts... tensors)
 Return an error if the passed tensor objects are not even. More...
 
template<typename... Ts>
arm_compute::Status error_on_tensors_not_subsampled (const char *function, const char *file, int line, const Format &format, const TensorShape &shape, const ITensor *tensor1, Ts... tensors)
 Return an error if the passed tensor objects are not sub-sampled. More...
 
template<typename... Ts>
arm_compute::Status error_on_mismatching_shapes (const char *function, const char *file, const int line, const ITensorInfo *tensor_info_1, const ITensorInfo *tensor_info_2, Ts... tensor_infos)
 Return an error if the passed two tensor infos have different shapes from the given dimension. More...
 
template<typename... Ts>
arm_compute::Status error_on_mismatching_shapes (const char *function, const char *file, const int line, const ITensor *tensor_1, const ITensor *tensor_2, Ts... tensors)
 Return an error if the passed two tensors have different shapes from the given dimension. More...
 
template<typename... Ts>
arm_compute::Status error_on_mismatching_shapes (const char *function, const char *file, const int line, unsigned int upper_dim, const ITensorInfo *tensor_info_1, const ITensorInfo *tensor_info_2, Ts... tensor_infos)
 Return an error if the passed two tensors have different shapes from the given dimension. More...
 
template<typename... Ts>
arm_compute::Status error_on_mismatching_shapes (const char *function, const char *file, const int line, unsigned int upper_dim, const ITensor *tensor_1, const ITensor *tensor_2, Ts... tensors)
 Return an error if the passed two tensors have different shapes from the given dimension. More...
 
template<typename... Ts>
arm_compute::Status error_on_mismatching_data_layouts (const char *function, const char *file, const int line, const ITensorInfo *tensor_info, Ts... tensor_infos)
 Return an error if the passed tensor infos have different data layouts. More...
 
template<typename... Ts>
arm_compute::Status error_on_mismatching_data_layouts (const char *function, const char *file, const int line, const ITensor *tensor, Ts... tensors)
 Return an error if the passed tensors have different data layouts. More...
 
template<typename... Ts>
arm_compute::Status error_on_mismatching_data_types (const char *function, const char *file, const int line, const ITensorInfo *tensor_info, Ts... tensor_infos)
 Return an error if the passed two tensor infos have different data types. More...
 
template<typename... Ts>
arm_compute::Status error_on_mismatching_data_types (const char *function, const char *file, const int line, const ITensor *tensor, Ts... tensors)
 Return an error if the passed two tensors have different data types. More...
 
template<typename... Ts>
arm_compute::Status error_on_mismatching_quantization_info (const char *function, const char *file, const int line, const ITensorInfo *tensor_info_1, const ITensorInfo *tensor_info_2, Ts... tensor_infos)
 Return an error if the passed tensor infos have different asymmetric quantized data types or different quantization info. More...
 
template<typename... Ts>
arm_compute::Status error_on_mismatching_quantization_info (const char *function, const char *file, const int line, const ITensor *tensor_1, const ITensor *tensor_2, Ts... tensors)
 Return an error if the passed tensor have different asymmetric quantized data types or different quantization info. More...
 
template<typename T , typename F , typename... Fs>
void error_on_format_not_in (const char *function, const char *file, const int line, const T *object, F &&format, Fs &&... formats)
 Throw an error if the format of the passed tensor/multi-image does not match any of the formats provided. More...
 
template<typename T , typename... Ts>
arm_compute::Status error_on_data_type_not_in (const char *function, const char *file, const int line, const ITensorInfo *tensor_info, T &&dt, Ts &&... dts)
 Return an error if the data type of the passed tensor info does not match any of the data types provided. More...
 
template<typename T , typename... Ts>
arm_compute::Status error_on_data_type_not_in (const char *function, const char *file, const int line, const ITensor *tensor, T &&dt, Ts &&... dts)
 Return an error if the data type of the passed tensor does not match any of the data types provided. More...
 
template<typename T , typename... Ts>
arm_compute::Status error_on_data_layout_not_in (const char *function, const char *file, const int line, const ITensorInfo *tensor_info, T &&dl, Ts &&... dls)
 Return an error if the data layout of the passed tensor info does not match any of the data layouts provided. More...
 
template<typename T , typename... Ts>
arm_compute::Status error_on_data_layout_not_in (const char *function, const char *file, const int line, const ITensor *tensor, T &&dl, Ts &&... dls)
 Return an error if the data layout of the passed tensor does not match any of the data layout provided. More...
 
template<typename T , typename... Ts>
arm_compute::Status error_on_data_type_channel_not_in (const char *function, const char *file, const int line, const ITensorInfo *tensor_info, size_t num_channels, T &&dt, Ts &&... dts)
 Return an error if the data type or the number of channels of the passed tensor info does not match any of the data types and number of channels provided. More...
 
template<typename T , typename... Ts>
arm_compute::Status error_on_data_type_channel_not_in (const char *function, const char *file, const int line, const ITensor *tensor, size_t num_channels, T &&dt, Ts &&... dts)
 Return an error if the data type or the number of channels of the passed tensor does not match any of the data types and number of channels provided. More...
 
arm_compute::Status error_on_unsupported_fp16 (const char *function, const char *file, const int line, const ITensorInfo *tensor_info, bool is_fp16_supported)
 Return an error if the data type of the passed tensor info is FP16 and FP16 extension is not supported by the device. More...
 
arm_compute::Status error_on_unsupported_fp16 (const char *function, const char *file, const int line, const ITensor *tensor, bool is_fp16_supported)
 Return an error if the data type of the passed tensor is FP16 and FP16 extension is not supported by the device. More...
 
arm_compute::Status error_on_tensor_not_2d (const char *function, const char *file, const int line, const ITensor *tensor)
 Return an error if the tensor is not 2D. More...
 
arm_compute::Status error_on_tensor_not_2d (const char *function, const char *file, const int line, const ITensorInfo *tensor)
 Return an error if the tensor info is not 2D. More...
 
template<typename T , typename... Ts>
arm_compute::Status error_on_channel_not_in (const char *function, const char *file, const int line, T cn, T &&channel, Ts &&... channels)
 Return an error if the channel is not in channels. More...
 
arm_compute::Status error_on_channel_not_in_known_format (const char *function, const char *file, const int line, Format fmt, Channel cn)
 Return an error if the channel is not in format. More...
 
arm_compute::Status error_on_invalid_multi_hog (const char *function, const char *file, const int line, const IMultiHOG *multi_hog)
 Return an error if the IMultiHOG container is invalid. More...
 
arm_compute::Status error_on_unconfigured_kernel (const char *function, const char *file, const int line, const IKernel *kernel)
 Return an error if the kernel is not configured. More...
 
arm_compute::Status error_on_invalid_subtensor (const char *function, const char *file, const int line, const TensorShape &parent_shape, const Coordinates &coords, const TensorShape &shape)
 Return an error if if the coordinates and shape of the subtensor are within the parent tensor. More...
 
arm_compute::Status error_on_invalid_subtensor_valid_region (const char *function, const char *file, const int line, const ValidRegion &parent_valid_region, const ValidRegion &valid_region)
 Return an error if the valid region of a subtensor is not inside the valid region of the parent tensor. More...
 
void swap (Window &lhs, Window &rhs)
 
Coordinates convert_window_coord_to_position (const Window &w, const Coordinates &offset)
 Convert an offset in window steps into absolute coordinates. More...
 
template<typename L >
WindowIterator< L > create_window_iterator (const Window &w, const Coordinates &start, const Coordinates &end, L &&lambda_function)
 Create a WindowIterator object. More...
 
arm_compute::DataType data_type_from_name (const std::string &name)
 Converts a string to a strong types enumeration DataType. More...
 
inline ::std::istream & operator>> (::std::istream &stream, arm_compute::DataType &data_type)
 Input Stream operator for DataType. More...
 
arm_compute::DataLayout data_layout_from_name (const std::string &name)
 Converts a string to a strong types enumeration DataLayout. More...
 
inline ::std::istream & operator>> (::std::istream &stream, arm_compute::DataLayout &data_layout)
 Input Stream operator for DataLayout. More...
 
std::tuple< cl::Context, cl::Device, cl_int > create_opencl_context_and_device ()
 This function creates an OpenCL context and a device. More...
 
CLTunerMode tuner_mode_from_name (const std::string &name)
 Converts a string to a strong types enumeration CLTunerMode. More...
 
inline ::std::istream & operator>> (::std::istream &stream, CLTunerMode &tuner_mode)
 Input Stream operator for CLTunerMode. More...
 
void get_cpu_configuration (CPUInfo &cpuinfo)
 This function will try to detect the CPU configuration on the system and will fill the cpuinfo object accordingly to reflect this. More...
 
unsigned int get_threads_hint ()
 Some systems have both big and small cores, this fuction computes the minimum number of cores that are exactly the same on the system. More...
 
const std::string & string_from_scheduler_type (Scheduler::Type t)
 Convert a Scheduler::Type into a string. More...
 
int32_t FloatFlip (float val)
 
float IFloatFlip (int32_t val)
 
std::pair< Status, Windowvalidate_and_configure_window (ITensorInfo *input, ITensorInfo *weights, ITensorInfo *biases, ITensorInfo *output, const PadStrideInfo &conv_info, unsigned int depth_multiplier, const Size2D &dilation)
 
template<typename T >
void run_reverse (const Window &window, const ITensor *input, const ITensor *axis, ITensor *output)
 
template<int lane>
uint8_t vget_lane (vec_8_byte_t< uint8_t > vec)
 
template<int lane>
uint8_t vget_lane (vec_16_byte_t< uint8_t > vec)
 
template<int lane>
int8_t vget_lane (vec_8_byte_t< int8_t > vec)
 
template<int lane>
int8_t vget_lane (vec_16_byte_t< int8_t > vec)
 
template<int lane>
uint16_t vget_lane (vec_8_byte_t< uint16_t > vec)
 
template<int lane>
uint16_t vget_lane (vec_16_byte_t< uint16_t > vec)
 
template<int lane>
int16_t vget_lane (vec_8_byte_t< int16_t > vec)
 
template<int lane>
int16_t vget_lane (vec_16_byte_t< int16_t > vec)
 
template<int lane>
uint32_t vget_lane (vec_8_byte_t< uint32_t > vec)
 
template<int lane>
uint32_t vget_lane (vec_16_byte_t< uint32_t > vec)
 
template<int lane>
int32_t vget_lane (vec_8_byte_t< int32_t > vec)
 
template<int lane>
int32_t vget_lane (vec_16_byte_t< int32_t > vec)
 
template<int lane>
float vget_lane (vec_8_byte_t< float > vec)
 
template<int lane>
float vget_lane (vec_16_byte_t< float > vec)
 
template<int lane>
float vget_lane (float32x4x4_t vec)
 
template<typename V >
constexpr size_t vec_size_of (const V &vec)
 
template<typename V >
vdup_n (elem_type_t< V > val)
 
template<typename V >
vld (const_ptr_t< elem_type_t< V >> ptr)
 
template<typename T >
sqadd (T a, T b)
 
template<typename T >
sqsub (T a, T b)
 
template<typename T >
sqmul (T a, T b)
 
template<>
vec_8_byte_t< uint8_t > vdup_n< vec_8_byte_t< uint8_t > > (uint8_t val)
 
template<>
vec_16_byte_t< uint8_t > vdup_n< vec_16_byte_t< uint8_t > > (uint8_t val)
 
template<>
vec_8_byte_t< uint8_t > vld< vec_8_byte_t< uint8_t > > (const_ptr_t< uint8_t > ptr)
 
template<>
vec_16_byte_t< uint8_t > vld< vec_16_byte_t< uint8_t > > (const_ptr_t< uint8_t > ptr)
 
void vst (ptr_t< uint8_t > ptr, vec_8_byte_t< uint8_t > vec)
 
void vst (ptr_t< uint8_t > ptr, vec_16_byte_t< uint8_t > vec)
 
vec_16_byte_t< uint8_t > vmax (vec_16_byte_t< uint8_t > a, vec_16_byte_t< uint8_t > b)
 
vec_8_byte_t< uint8_t > vpmax (vec_8_byte_t< uint8_t > a, vec_8_byte_t< uint8_t > b)
 
vec_8_byte_t< uint8_t > vget_low (vec_16_byte_t< uint8_t > vec)
 
vec_8_byte_t< uint8_t > vget_high (vec_16_byte_t< uint8_t > vec)
 
template<>
vec_8_byte_t< int8_t > vdup_n< vec_8_byte_t< int8_t > > (int8_t val)
 
template<>
vec_16_byte_t< int8_t > vdup_n< vec_16_byte_t< int8_t > > (int8_t val)
 
template<>
vec_8_byte_t< int8_t > vld< vec_8_byte_t< int8_t > > (const_ptr_t< int8_t > ptr)
 
template<>
vec_16_byte_t< int8_t > vld< vec_16_byte_t< int8_t > > (const_ptr_t< int8_t > ptr)
 
void vst (ptr_t< int8_t > ptr, vec_8_byte_t< int8_t > vec)
 
void vst (ptr_t< int8_t > ptr, vec_16_byte_t< int8_t > vec)
 
vec_16_byte_t< int8_t > vmax (vec_16_byte_t< int8_t > a, vec_16_byte_t< int8_t > b)
 
vec_8_byte_t< int8_t > vpmax (vec_8_byte_t< int8_t > a, vec_8_byte_t< int8_t > b)
 
vec_8_byte_t< int8_t > vget_low (vec_16_byte_t< int8_t > vec)
 
vec_8_byte_t< int8_t > vget_high (vec_16_byte_t< int8_t > vec)
 
template<>
vec_8_byte_t< uint16_t > vdup_n< vec_8_byte_t< uint16_t > > (uint16_t val)
 
template<>
vec_16_byte_t< uint16_t > vdup_n< vec_16_byte_t< uint16_t > > (uint16_t val)
 
template<>
vec_8_byte_t< uint16_t > vld< vec_8_byte_t< uint16_t > > (const_ptr_t< uint16_t > ptr)
 
template<>
vec_16_byte_t< uint16_t > vld< vec_16_byte_t< uint16_t > > (const_ptr_t< uint16_t > ptr)
 
void vst (ptr_t< uint16_t > ptr, vec_8_byte_t< uint16_t > vec)
 
void vst (ptr_t< uint16_t > ptr, vec_16_byte_t< uint16_t > vec)
 
vec_16_byte_t< uint16_t > vmax (vec_16_byte_t< uint16_t > a, vec_16_byte_t< uint16_t > b)
 
vec_8_byte_t< uint16_t > vpmax (vec_8_byte_t< uint16_t > a, vec_8_byte_t< uint16_t > b)
 
vec_8_byte_t< uint16_t > vget_low (vec_16_byte_t< uint16_t > vec)
 
vec_8_byte_t< uint16_t > vget_high (vec_16_byte_t< uint16_t > vec)
 
template<>
vec_8_byte_t< int16_t > vdup_n< vec_8_byte_t< int16_t > > (int16_t val)
 
template<>
vec_16_byte_t< int16_t > vdup_n< vec_16_byte_t< int16_t > > (int16_t val)
 
template<>
vec_8_byte_t< int16_t > vld< vec_8_byte_t< int16_t > > (const_ptr_t< int16_t > ptr)
 
template<>
vec_16_byte_t< int16_t > vld< vec_16_byte_t< int16_t > > (const_ptr_t< int16_t > ptr)
 
void vst (ptr_t< int16_t > ptr, vec_8_byte_t< int16_t > vec)
 
void vst (ptr_t< int16_t > ptr, vec_16_byte_t< int16_t > vec)
 
vec_16_byte_t< int16_t > vmax (vec_16_byte_t< int16_t > a, vec_16_byte_t< int16_t > b)
 
vec_8_byte_t< int16_t > vpmax (vec_8_byte_t< int16_t > a, vec_8_byte_t< int16_t > b)
 
vec_8_byte_t< int16_t > vget_low (vec_16_byte_t< int16_t > vec)
 
vec_8_byte_t< int16_t > vget_high (vec_16_byte_t< int16_t > vec)
 
template<>
vec_8_byte_t< uint32_t > vdup_n< vec_8_byte_t< uint32_t > > (uint32_t val)
 
template<>
vec_16_byte_t< uint32_t > vdup_n< vec_16_byte_t< uint32_t > > (uint32_t val)
 
template<>
vec_8_byte_t< uint32_t > vld< vec_8_byte_t< uint32_t > > (const_ptr_t< uint32_t > ptr)
 
template<>
vec_16_byte_t< uint32_t > vld< vec_16_byte_t< uint32_t > > (const_ptr_t< uint32_t > ptr)
 
void vst (ptr_t< uint32_t > ptr, vec_8_byte_t< uint32_t > vec)
 
void vst (ptr_t< uint32_t > ptr, vec_16_byte_t< uint32_t > vec)
 
vec_16_byte_t< uint32_t > vmax (vec_16_byte_t< uint32_t > a, vec_16_byte_t< uint32_t > b)
 
vec_8_byte_t< uint32_t > vpmax (vec_8_byte_t< uint32_t > a, vec_8_byte_t< uint32_t > b)
 
vec_8_byte_t< uint32_t > vget_low (vec_16_byte_t< uint32_t > vec)
 
vec_8_byte_t< uint32_t > vget_high (vec_16_byte_t< uint32_t > vec)
 
template<>
vec_8_byte_t< int32_t > vdup_n< vec_8_byte_t< int32_t > > (int32_t val)
 
template<>
vec_16_byte_t< int32_t > vdup_n< vec_16_byte_t< int32_t > > (int32_t val)
 
template<>
vec_8_byte_t< int32_t > vld< vec_8_byte_t< int32_t > > (const_ptr_t< int32_t > ptr)
 
template<>
vec_16_byte_t< int32_t > vld< vec_16_byte_t< int32_t > > (const_ptr_t< int32_t > ptr)
 
void vst (ptr_t< int32_t > ptr, vec_8_byte_t< int32_t > vec)
 
void vst (ptr_t< int32_t > ptr, vec_16_byte_t< int32_t > vec)
 
vec_16_byte_t< int32_t > vmax (vec_16_byte_t< int32_t > a, vec_16_byte_t< int32_t > b)
 
vec_8_byte_t< int32_t > vpmax (vec_8_byte_t< int32_t > a, vec_8_byte_t< int32_t > b)
 
vec_8_byte_t< int32_t > vget_low (vec_16_byte_t< int32_t > vec)
 
vec_8_byte_t< int32_t > vget_high (vec_16_byte_t< int32_t > vec)
 
template<>
vec_8_byte_t< float > vdup_n< vec_8_byte_t< float > > (float val)
 
template<>
vec_16_byte_t< float > vdup_n< vec_16_byte_t< float > > (float val)
 
template<>
vec_8_byte_t< float > vld< vec_8_byte_t< float > > (const_ptr_t< float > ptr)
 
template<>
vec_16_byte_t< float > vld< vec_16_byte_t< float > > (const_ptr_t< float > ptr)
 
void vst (ptr_t< float > ptr, vec_8_byte_t< float > vec)
 
void vst (ptr_t< float > ptr, vec_16_byte_t< float > vec)
 
vec_16_byte_t< float > vmax (vec_16_byte_t< float > a, vec_16_byte_t< float > b)
 
vec_8_byte_t< float > vpmax (vec_8_byte_t< float > a, vec_8_byte_t< float > b)
 
vec_8_byte_t< float > vget_low (vec_16_byte_t< float > vec)
 
vec_8_byte_t< float > vget_high (vec_16_byte_t< float > vec)
 
vec_8_byte_t< float > vadd (vec_8_byte_t< float > a, vec_8_byte_t< float > b)
 
vec_16_byte_t< float > vadd (vec_16_byte_t< float > a, vec_16_byte_t< float > b)
 
vec_16_byte_t< float > vsub (vec_16_byte_t< float > a, vec_16_byte_t< float > b)
 
vec_16_byte_t< float > vmul_n (vec_16_byte_t< float > vec, float val)
 
template<typename VO , typename VI >
VO vcvt (VI vec)
 
template<>
float32x4x4_t vcvt< float32x4x4_t > (uint8x16_t vec)
 
template<>
uint8x16_t vcvt< uint8x16_t > (float32x4x4_t vec)
 
float32x4x4_t vexp (float32x4x4_t vec)
 
float32x4_t vexp (const float32x4_t &vec)
 
template<>
float32x4x4_t vdup_n< float32x4x4_t > (float val)
 
float32x4x4_t vmul_n (float32x4x4_t vec, float val)
 
float32x4x4_t vadd (float32x4x4_t a, float32x4x4_t b)
 
inline ::std::ostream & operator<< (::std::ostream &os, const GradientDimension &dim)
 Formatted output of the GradientDimension type. More...
 
std::string to_string (const arm_compute::GradientDimension &type)
 Formatted output of the GradientDimension type. More...
 
inline ::std::istream & operator>> (::std::istream &is, BorderMode &mode)
 Formatted input of the BorderMode type. More...
 
template<typename T >
std::string to_string_if_not_null (T *arg)
 Formatted output if arg is not null. More...
 
template<typename T >
inline ::std::ostream & operator<< (::std::ostream &os, const Dimensions< T > &dimensions)
 Formatted output of the Dimensions type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const NonLinearFilterFunction &function)
 Formatted output of the NonLinearFilterFunction type. More...
 
std::string to_string (const NonLinearFilterFunction &function)
 Formatted output of the NonLinearFilterFunction type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const MatrixPattern &pattern)
 Formatted output of the MatrixPattern type. More...
 
std::string to_string (const MatrixPattern &pattern)
 Formatted output of the MatrixPattern type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const RoundingPolicy &rounding_policy)
 Formatted output of the RoundingPolicy type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const WeightsInfo &weights_info)
 Formatted output of the WeightsInfo type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const ROIPoolingLayerInfo &pool_info)
 Formatted output of the ROIPoolingInfo type. More...
 
std::string to_string (const ROIPoolingLayerInfo &pool_info)
 Formatted output of the ROIPoolingInfo type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const BoundingBoxTransformInfo &bbox_info)
 Formatted output of the BoundingBoxTransformInfo type. More...
 
std::string to_string (const BoundingBoxTransformInfo &bbox_info)
 Formatted output of the BoundingBoxTransformInfo type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const ComputeAnchorsInfo &anchors_info)
 Formatted output of the ComputeAnchorsInfo type. More...
 
std::string to_string (const ComputeAnchorsInfo &anchors_info)
 Formatted output of the ComputeAnchorsInfo type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const GenerateProposalsInfo &proposals_info)
 Formatted output of the GenerateProposalsInfo type. More...
 
std::string to_string (const GenerateProposalsInfo &proposals_info)
 Formatted output of the GenerateProposalsInfo type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const QuantizationInfo &qinfo)
 Formatted output of the QuantizationInfo type. More...
 
std::string to_string (const QuantizationInfo &quantization_info)
 Formatted output of the QuantizationInfo type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const ActivationLayerInfo::ActivationFunction &act_function)
 Formatted output of the activation function type. More...
 
std::string to_string (const arm_compute::ActivationLayerInfo &info)
 Formatted output of the activation function info type. More...
 
std::string to_string (const arm_compute::ActivationLayerInfo::ActivationFunction &function)
 Formatted output of the activation function type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const NormType &norm_type)
 Formatted output of the NormType type. More...
 
std::string to_string (const arm_compute::NormalizationLayerInfo &info)
 Formatted output of NormalizationLayerInfo. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const NormalizationLayerInfo &info)
 Formatted output of NormalizationLayerInfo. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const PoolingType &pool_type)
 Formatted output of the PoolingType type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const PoolingLayerInfo &info)
 Formatted output of PoolingLayerInfo. More...
 
std::string to_string (const RoundingPolicy &rounding_policy)
 Formatted output of RoundingPolicy. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const DataLayout &data_layout)
 [Print DataLayout type] More...
 
std::string to_string (const arm_compute::DataLayout &data_layout)
 Formatted output of the DataLayout type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const DataLayoutDimension &data_layout_dim)
 [Print DataLayout type] More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const DataType &data_type)
 Formatted output of the DataType type. More...
 
std::string to_string (const arm_compute::DataType &data_type)
 Formatted output of the DataType type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const Format &format)
 Formatted output of the Format type. More...
 
std::string to_string (const Format &format)
 Formatted output of the Format type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const Channel &channel)
 Formatted output of the Channel type. More...
 
std::string to_string (const Channel &channel)
 Formatted output of the Channel type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const BorderMode &mode)
 Formatted output of the BorderMode type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const BorderSize &border)
 Formatted output of the BorderSize type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const PaddingList &padding)
 Formatted output of the PaddingList type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const Multiples &multiples)
 Formatted output of the Multiples type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const InterpolationPolicy &policy)
 Formatted output of the InterpolationPolicy type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const SamplingPolicy &policy)
 Formatted output of the SamplingPolicy type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const TensorInfo &info)
 Formatted output of the TensorInfo type. More...
 
std::string to_string (const TensorInfo &info)
 Formatted output of the TensorInfo type. More...
 
template<typename T >
std::string to_string (const Dimensions< T > &dimensions)
 Formatted output of the Dimensions type. More...
 
std::string to_string (const Strides &stride)
 Formatted output of the Strides type. More...
 
std::string to_string (const TensorShape &shape)
 Formatted output of the TensorShape type. More...
 
std::string to_string (const Coordinates &coord)
 Formatted output of the Coordinates type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const GEMMReshapeInfo &info)
 Formatted output of the GEMMReshapeInfo type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const GEMMInfo &info)
 Formatted output of the GEMMInfo type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const Window::Dimension &dim)
 Formatted output of the Window::Dimension type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const Window &win)
 Formatted output of the Window type. More...
 
std::string to_string (const WeightsInfo &info)
 Formatted output of the WeightsInfo type. More...
 
std::string to_string (const GEMMReshapeInfo &info)
 Formatted output of the GEMMReshapeInfo type. More...
 
std::string to_string (const GEMMInfo &info)
 Formatted output of the GEMMInfo type. More...
 
std::string to_string (const Window::Dimension &dim)
 Formatted output of the Window::Dimension type. More...
 
std::string to_string (const Window &win)
 Formatted output of the Window type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const Rectangle &rect)
 Formatted output of the Rectangle type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const PaddingMode &mode)
 Formatted output of the PaddingMode type. More...
 
std::string to_string (const PaddingMode &mode)
 Formatted output of the PaddingMode type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const PadStrideInfo &pad_stride_info)
 Formatted output of the PadStrideInfo type. More...
 
std::string to_string (const PadStrideInfo &pad_stride_info)
 Formatted output of the PadStrideInfo type. More...
 
std::string to_string (const BorderMode &mode)
 Formatted output of the BorderMode type. More...
 
std::string to_string (const BorderSize &border)
 Formatted output of the BorderSize type. More...
 
std::string to_string (const PaddingList &padding)
 Formatted output of the PaddingList type. More...
 
std::string to_string (const Multiples &multiples)
 Formatted output of the Multiples type. More...
 
std::string to_string (const InterpolationPolicy &policy)
 Formatted output of the InterpolationPolicy type. More...
 
std::string to_string (const SamplingPolicy &policy)
 Formatted output of the SamplingPolicy type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const ConvertPolicy &policy)
 Formatted output of the ConvertPolicy type. More...
 
std::string to_string (const ConvertPolicy &policy)
 
inline ::std::ostream & operator<< (::std::ostream &os, const ArithmeticOperation &op)
 Formatted output of the ArithmeticOperation type. More...
 
std::string to_string (const ArithmeticOperation &op)
 Formatted output of the Arithmetic Operation. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const ReductionOperation &op)
 Formatted output of the Reduction Operations. More...
 
std::string to_string (const ReductionOperation &op)
 Formatted output of the Reduction Operations. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const ComparisonOperation &op)
 Formatted output of the Comparison Operations. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const ElementWiseUnary &op)
 Formatted output of the Elementwise unary Operations. More...
 
std::string to_string (const ComparisonOperation &op)
 Formatted output of the Comparison Operations. More...
 
std::string to_string (const ElementWiseUnary &op)
 Formatted output of the Elementwise unary Operations. More...
 
std::string to_string (const NormType &type)
 Formatted output of the Norm Type. More...
 
std::string to_string (const PoolingType &type)
 Formatted output of the Pooling Type. More...
 
std::string to_string (const PoolingLayerInfo &info)
 Formatted output of the Pooling Layer Info. More...
 
std::string to_string (const PriorBoxLayerInfo &info)
 Formatted output of the PriorBoxLayerInfo. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const KeyPoint &point)
 Formatted output of the KeyPoint type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const PhaseType &phase_type)
 Formatted output of the PhaseType type. More...
 
std::string to_string (const arm_compute::PhaseType &type)
 Formatted output of the PhaseType type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const MagnitudeType &magnitude_type)
 Formatted output of the MagnitudeType type. More...
 
std::string to_string (const arm_compute::MagnitudeType &type)
 Formatted output of the MagnitudeType type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const HOGNormType &norm_type)
 Formatted output of the HOGNormType type. More...
 
std::string to_string (const HOGNormType &type)
 Formatted output of the HOGNormType type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const Size2D &size)
 Formatted output of the Size2D type. More...
 
std::string to_string (const Size2D &type)
 Formatted output of the Size2D type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const HOGInfo &hog_info)
 Formatted output of the HOGInfo type. More...
 
std::string to_string (const HOGInfo &type)
 Formatted output of the HOGInfo type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const ConvolutionMethod &conv_method)
 Formatted output of the ConvolutionMethod type. More...
 
std::string to_string (const ConvolutionMethod &conv_method)
 Formatted output of the ConvolutionMethod type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const GPUTarget &gpu_target)
 Formatted output of the GPUTarget type. More...
 
std::string to_string (const GPUTarget &gpu_target)
 Formatted output of the GPUTarget type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const DetectionWindow &detection_window)
 Formatted output of the DetectionWindow type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const DetectionOutputLayerCodeType &detection_code)
 Formatted output of the DetectionOutputLayerCodeType type. More...
 
std::string to_string (const DetectionOutputLayerCodeType &detection_code)
 Formatted output of the DetectionOutputLayerCodeType type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const DetectionOutputLayerInfo &detection_info)
 Formatted output of the DetectionOutputLayerInfo type. More...
 
std::string to_string (const DetectionOutputLayerInfo &detection_info)
 Formatted output of the DetectionOutputLayerInfo type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const DetectionPostProcessLayerInfo &detection_info)
 Formatted output of the DetectionPostProcessLayerInfo type. More...
 
std::string to_string (const DetectionPostProcessLayerInfo &detection_info)
 Formatted output of the DetectionPostProcessLayerInfo type. More...
 
std::string to_string (const DetectionWindow &detection_window)
 Formatted output of the DetectionWindow type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const Termination &termination)
 Formatted output of the Termination type. More...
 
std::string to_string (const Termination &termination)
 Formatted output of the Termination type. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const CPUModel &cpu_model)
 Formatted output of the CPUModel type. More...
 
std::string to_string (const CPUModel &cpu_model)
 Formatted output of the CPUModel type. More...
 
template<typename T >
inline ::std::ostream & operator<< (::std::ostream &os, const std::vector< T > &args)
 Formatted output of a vector of objects. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const PriorBoxLayerInfo &info)
 Formatted output of PriorBoxLayerInfo. More...
 
template<typename T >
std::string to_string (const std::vector< T > &args)
 Formatted output of a vector of objects. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const WinogradInfo &info)
 Formatted output of the WinogradInfo type. More...
 
std::string to_string (const WinogradInfo &type)
 
template<typename T >
std::string to_string (const T &val)
 Fallback method: try to use std::to_string: More...
 
std::string to_string (const CLTunerMode val)
 Convert a CLTunerMode value to a string. More...
 
inline ::std::ostream & operator<< (::std::ostream &os, const CLTunerMode &val)
 [Print CLTunerMode type] More...
 

Variables

constexpr size_t MAX_DIMS = 6
 Constant value used to indicate maximum dimensions of a Window, TensorShape and Coordinates. More...
 
const std::array< float32x4_t, 8 > exp_tab
 Exponent polynomial coefficients. More...
 
const std::array< float32x4_t, 8 > log_tab
 Logarithm polynomial coefficients. More...
 
constexpr float te_sin_coeff2 = 0.166666666666f
 Sin polynomial coefficients. More...
 
constexpr float te_sin_coeff3 = 0.05f
 
constexpr float te_sin_coeff4 = 0.023809523810f
 
constexpr float te_sin_coeff5 = 0.013888888889f
 
constexpr uint8_t CONSTANT_BORDER_VALUE = 199
 Constant value of the border pixels when using BorderMode::CONSTANT. More...
 
constexpr float SCALE_PYRAMID_HALF = 0.5f
 Constant value used to indicate a half-scale pyramid. More...
 
constexpr float SCALE_PYRAMID_ORB = 8.408964152537146130583778358414e-01
 Constant value used to indicate a ORB scaled pyramid. More...
 
constexpr unsigned int num_num_elems_processed_per_iteration = 16
 

Detailed Description

Copyright (c) 2017-2018 ARM Limited.

This file contains all available output stages for GEMMLowp on NEON.

This file contains all available output stages for GEMMLowp on OpenCL.

Copyright (c) 2018 ARM Limited.

Copyright (c) 2018-2019 ARM Limited.

SPDX-License-Identifier: MIT

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

In gemmlowp, the "output stage" is the process that takes a final int32 accumulator value (the output of CLGEMMLowpMatrixMultiplyCore), and processes it to obtain the final ASYMM8 value.

More information about the GEMMLowp output stage can be found at https://github.com/google/gemmlowp/blob/master/doc/output.md

In gemmlowp, the "output stage" is the process that takes a final int32 accumulator value (the output of NEGEMMLowpMatrixMultiplyCore), and processes it to obtain the final ASYMM8 value.

More information about the GEMMLowp output stage can be found at https://github.com/google/gemmlowp/blob/master/doc/output.md

Typedef Documentation

◆ BBox

using BBox = std::array<float, 4>

Definition at line 948 of file Types.h.

◆ BiStrides

Bidirectional strides.

Definition at line 49 of file Types.h.

◆ CLCoefficientTableArray

OpenCL Array of Coefficient Tables.

Definition at line 49 of file CLOpticalFlow.h.

◆ CLConvolution3x3Kernel

Interface for the kernel which applies a 3x3 convolution to a tensor.

Definition at line 70 of file CLConvolutionKernel.h.

◆ CLConvolution5x5

Basic function to run 5x5 convolution.

Definition at line 102 of file CLConvolution.h.

◆ CLConvolution5x5Kernel

Interface for the kernel which applies a 5x5 convolution to a tensor.

Definition at line 72 of file CLConvolutionKernel.h.

◆ CLConvolution7x7

Basic function to run 7x7 convolution.

Definition at line 104 of file CLConvolution.h.

◆ CLConvolution7x7Kernel

Interface for the kernel which applies a 7x7 convolution to a tensor.

Definition at line 74 of file CLConvolutionKernel.h.

◆ CLConvolution9x9

Basic function to run 9x9 convolution.

Definition at line 106 of file CLConvolution.h.

◆ CLConvolution9x9Kernel

Interface for the kernel which applies a 9x9 convolution to a tensor.

Definition at line 76 of file CLConvolutionKernel.h.

◆ CLCoordinates2DArray

OpenCL Array of 2D Coordinates.

Definition at line 109 of file CLArray.h.