NN Namespace Reference

NN Neural Network Library containing Tensor and NeuralNet classes for inference. More...

Namespaces
namespace	Util
	Namespace for utility functions related to neural networks and tensors.

Classes
struct	is_std_array
	Checks if a type is a std::array. More...
struct	is_std_array< std::array< U, N > >
struct	map_array_traits
	Traits for mapping a function over a std::array. More...
class	NeuralError
	Represents an error that occurred in the neural network module and contains the error type and message. More...
class	NeuralNet
	Abstract base class for neural networks. More...
class	NeuralNetFactory
	Factory class for creating NeuralNet instances. This abstracts from the underlying inference engine (e.g., ONNX Runtime). More...
class	OrtNeuralNet
	A class that implements the NeuralNet interface using ONNX Runtime. More...
struct	ReduceArgMax
	Reduces a Tensor by finding the index of the maximum value along a specified axis. More...
struct	ReduceArgMin
	Reduces a Tensor by finding the index of the minimum value along a specified axis. More...
struct	ReduceMax
	Reduces a Tensor by computing the maximum value along a specified axis. More...
struct	ReduceMin
	Reduces a Tensor by computing the minimum value along a specified axis. More...
struct	ReduceSum
	Reduces a Tensor by summing its elements along a specified axis. More...
class	Tensor
	A Tensor represents a N-dimensional array containing elements of the same type. Can be used as input and output for inference. More...

Typedefs
using	NeuralNetPtr = std::shared_ptr< NeuralNet >
	Smart pointer type for managing NeuralNet instances.
using	Shape = std::vector< int64_t >
	Shape of a N-dimensional Tensor represented as the size in each dimension. Can be -1 in case of dynamic dimensions.
template<typename T >
using	decay_t = typename std::decay< T >::type
	Remove cv/ref qualifiers and decay to check std::array<T, N>

Enumerations
enum class	ErrorCode
	Error codes for the neural network module. More...
enum class	TensorType
	TensorType encapsulates the supported data types of tensor elements. The supported types are: More...

Functions
int64_t	shapeNumElements (const Shape &shape)
	Calculates the number of elements from a given Shape.
int64_t	shapeToStride (const Shape &shape, uint64_t axis)
	Calculates the stride to iterate elements in a given axis.
std::string	shapeToString (const Shape &shape)
	Creates a human-readable string from the given shape.
constexpr const char *	toString (TensorType type)
	Convert the TensorType to a human-readable string.

Detailed Description

NN Neural Network Library containing Tensor and NeuralNet classes for inference.

Contains the Tensor class for representing N-dimensional arrays with various data types. The Tensor class supports operations like reshaping, mapping, reducing, and converting to/from cv::Mat.

A factory handles model loading and abstracts the underlying inference engine (e.g., ONNX Runtime).

Author

Dominik Wüst (domin.nosp@m.ik.w.nosp@m.uest@.nosp@m.iosb.nosp@m..frau.nosp@m.nhof.nosp@m.er.de)

Date

May 2025

Function Documentation

shapeNumElements()

int64_t NN::shapeNumElements

(

const Shape &

shape

)

Calculates the number of elements from a given Shape.

Returns

shape[0] * shape[1] * ... * shape[N-1] 

Parameters

shape

The shape object.

Returns

int64_t The number of elements in a Tensor with that shape.

shapeToStride()

int64_t NN::shapeToStride	(	const Shape &	shape,
		uint64_t	axis
	)

Calculates the stride to iterate elements in a given axis.

Parameters

shape	The shape of the tensor to iterate.
axis	The axis to iterate. This must be in range [0,shape.size()-1]!

Returns

int64_t the stride (stepsize) to iterate the given axis.

shapeToString()

std::string NN::shapeToString

(

const Shape &

shape

)

Creates a human-readable string from the given shape.

Parameters

shape

The shape to convert.

Returns

std::string A string with the shapes elements in printable form.

toString()

constexpr const char * NN::toString ( TensorType  type )

constexpr

Convert the TensorType to a human-readable string.

Parameters

type	The TensorType to convert.

Returns

constexpr const char* Returns a string representation of the type.