v18.01のリリースノート
引用
Various bug fixes
Added some of the missing validate() methods
Added arm_compute::CLDeconvolutionLayerUpsampleKernel / arm_compute::CLDeconvolutionLayer arm_compute::CLDeconvolutionLayerUpsample
Added arm_compute::CLPermuteKernel / arm_compute::CLPermute
Added method to clean the programs cache in the CL Kernel library.
Added arm_compute::GCArithmeticAdditionKernel / arm_compute::GCArithmeticAddition
Added arm_compute::GCDepthwiseConvolutionLayer3x3Kernel / arm_compute::GCDepthwiseConvolutionLayer3x3
Added arm_compute::GCNormalizePlanarYUVLayerKernel / arm_compute::GCNormalizePlanarYUVLayer
Added arm_compute::GCScaleKernel / arm_compute::GCScale
Added arm_compute::GCWeightsReshapeKernel / arm_compute::GCConvolutionLayer
Added FP16 support to the following GLES compute kernels:
arm_compute::GCCol2ImKernel
arm_compute::GCGEMMInterleave4x4Kernel
arm_compute::GCGEMMTranspose1xWKernel
arm_compute::GCIm2ColKernel
Refactored NEON Winograd (arm_compute::NEWinogradLayerKernel)
Added arm_compute::NEDirectConvolutionLayerOutputStageKernel
Added QASYMM8 support to the following NEON kernels:
arm_compute::NEDepthwiseConvolutionLayer3x3Kernel
arm_compute::NEFillBorderKernel
arm_compute::NEPoolingLayerKernel
Added new examples:
graph_cl_mobilenet_qasymm8.cpp
graph_inception_v3.cpp
gc_dc.cpp
More tests added to both validation and benchmarking suites.
