23.02
|
Go to the documentation of this file.
12 class ScopedTensorHandle;
38 std::vector<TensorShape>
InferOutputShapes(
const std::vector<TensorShape>& inputShapes)
const override;
virtual std::unique_ptr< IWorkload > CreateWorkload(const IWorkloadFactory &factory) const override
Makes a workload for the FullyConnected type.
A FullyConnectedDescriptor for the FullyConnectedLayer.
FullyConnectedLayer(const FullyConnectedDescriptor ¶m, const char *name)
Constructor to create a FullyConnectedLayer.
std::vector< std::reference_wrapper< std::shared_ptr< ConstTensorHandle > >> ConstantTensors
~FullyConnectedLayer()=default
Default destructor.
void ReleaseConstantData() override
This layer does not have any data stored, weights and bias are now stored in constant layers.
Copyright (c) 2021 ARM Limited and Contributors.
This layer represents a fully connected operation.
void ExecuteStrategy(IStrategy &strategy) const override
Apply a visitor to this layer.
FullyConnectedLayer * Clone(Graph &graph) const override
Creates a dynamically-allocated copy of this layer.
std::vector< TensorShape > InferOutputShapes(const std::vector< TensorShape > &inputShapes) const override
By default returns inputShapes if the number of inputs are equal to number of outputs,...
ConstantTensors GetConstantTensorsByRef() override
Retrieve the handles to the constant values stored by the layer.
void ValidateTensorShapesFromInputs() override
Check if the input tensor shape(s) will lead to a valid configuration of FullyConnectedLayer.