24.08
|
#include <GpuFsaLayerSupport.hpp>
Public Member Functions | |
bool | IsLayerSupported (const LayerType &type, const std::vector< TensorInfo > &infos, const BaseDescriptor &descriptor, const Optional< LstmInputParamsInfo > &lstmParamsInfo, const Optional< QuantizedLstmInputParamsInfo > &, Optional< std::string & > reasonIfUnsupported) const override |
Default implementation of the ILayerSupport interface, Backends should implement this as a switch statement for each of their LayerTypes calling their specific backend implementation of IsXXXLayerSupported. More... | |
Additional Inherited Members | |
Protected Member Functions inherited from ILayerSupport | |
ILayerSupport () | |
virtual | ~ILayerSupport () |
Definition at line 13 of file GpuFsaLayerSupport.hpp.
|
overridevirtual |
Default implementation of the ILayerSupport interface, Backends should implement this as a switch statement for each of their LayerTypes calling their specific backend implementation of IsXXXLayerSupported.
Reimplemented from ILayerSupport.
Definition at line 71 of file GpuFsaLayerSupport.cpp.
References armnn::Activation, armnn::BatchMatMul, armnn::Cast, armnn::Constant, armnn::Convolution2d, armnn::DepthwiseConvolution2d, armnn::ElementwiseBinary, FORWARD_LAYER_VALIDATE_FUNC, armnn::GpuFsaActivationValidate(), armnn::GpuFsaBatchMatMulValidate(), armnn::GpuFsaCastValidate(), armnn::GpuFsaConvolution2dValidate(), armnn::GpuFsaDepthwiseConvolution2dValidate(), armnn::GpuFsaElementwiseBinaryValidate(), armnn::GpuFsaPooling2dValidate(), armnn::GpuFsaReshapeValidate(), armnn::GpuFsaResizeValidate(), armnn::GpuFsaSoftmaxValidate(), armnn::IgnoreUnused(), armnn::Input, armnn::IsGpuFsaBackendSupported(), armnn::Output, armnn::Pooling2d, armnn::Reshape, armnn::Resize, and armnn::Softmax.