24.08
|
Go to the documentation of this file.
12 class ScopedTensorHandle;
21 std::shared_ptr<ConstTensorHandle>
m_Bias;
41 std::vector<TensorShape>
InferOutputShapes(
const std::vector<TensorShape>& inputShapes)
const override;
void ValidateTensorShapesFromInputs() override
Check if the input tensor shape(s) will lead to a valid configuration of TransposeConvolution2dLayer.
ImmutableConstantTensors GetConstantTensorsByRef() const override
Retrieve the handles to the constant values stored by the layer.
This layer represents a 2D transpose convolution operation.
~TransposeConvolution2dLayer()=default
Default destructor.
std::vector< std::reference_wrapper< const std::shared_ptr< ConstTensorHandle > >> ImmutableConstantTensors
std::shared_ptr< ConstTensorHandle > m_Bias
A unique pointer to store bias values.
virtual std::unique_ptr< IWorkload > CreateWorkload(const IWorkloadFactory &factory) const override
Makes a workload for the TransposeConvolution2d type.
std::vector< TensorShape > InferOutputShapes(const std::vector< TensorShape > &inputShapes) const override
Infers the output shapes from given input shapes and layer properties.
TransposeConvolution2dLayer(const TransposeConvolution2dDescriptor ¶m, const char *name)
Constructor to create a TransposeConvolution2dLayer.
std::shared_ptr< ConstTensorHandle > m_Weight
A unique pointer to store weight values.
void ExecuteStrategy(IStrategy &strategy) const override
Apply a visitor to this layer.
Copyright (c) 2021 ARM Limited and Contributors.
A TransposeConvolution2dDescriptor for the TransposeConvolution2dLayer.
TransposeConvolution2dLayer * Clone(Graph &graph) const override
Creates a dynamically-allocated copy of this layer.