nvneural::TensorDimension Struct Reference

TensorDimension describes the dimensions of a four-dimensional image tensor. More...

#include <nvneural/CoreTypes.h>

Public Member Functions
std::size_t	elementCount () const
	Returns the number of elements in a tensor with these dimensions. More...
bool	operator!= (const TensorDimension &other) const
	Inequality operator.
TensorDimension &	operator= (const TensorDimension &copyFrom)
	Copy assignment operator.
bool	operator== (const TensorDimension &other) const
	Equality operator.
	TensorDimension ()=default
	Default constructor. Initializes all members to zero.
	TensorDimension (const TensorDimension &copyFrom)
	Copy constructor.
	TensorDimension (size_t N, size_t C, size_t H, size_t W)
	Direct constructor, specifying all members at once.
	~TensorDimension ()=default
	Trivial destructor.

Public Attributes
size_t	c = 0
	C dimension (channel)
size_t	h = 0
	H dimension (height)
size_t	n = 0
	N dimension (batch)
size_t	w = 0
	W dimension (width)

Detailed Description

TensorDimension describes the dimensions of a four-dimensional image tensor.

Its member functions are provided for convenience and may change implementations between versions of the nvneural SDK, but the binary layout of this type will remain constant.

Member Function Documentation

elementCount()

std::size_t nvneural::TensorDimension::elementCount ( ) const

inline

Returns the number of elements in a tensor with these dimensions.

This can be used to compute buffer size by multiplying by sizeof(element).

---

The documentation for this struct was generated from the following file:

• Core/Inc/nvneural/CoreTypes.h