24.02
|
Go to the documentation of this file.
31 std::vector<TensorShape>
InferOutputShapes(
const std::vector<TensorShape>& inputShapes)
const override;
ReduceLayer(const ReduceDescriptor ¶m, const char *name)
Constructor to create a ReduceLayer.
void ValidateTensorShapesFromInputs() override
Check if the input tensor shape(s) will lead to a valid configuration of ReduceLayer.
~ReduceLayer()=default
Default destructor.
This layer represents a reduction operation.
ReduceLayer * Clone(Graph &graph) const override
Creates a dynamically-allocated copy of this layer.
Copyright (c) 2021 ARM Limited and Contributors.
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,...
void ExecuteStrategy(IStrategy &strategy) const override
Apply a visitor to this layer.
A ReduceDescriptor for the REDUCE operators.
virtual std::unique_ptr< IWorkload > CreateWorkload(const IWorkloadFactory &factory) const override
Makes a workload for the Reduce type.