TensorRT  7.2.1.6
NVIDIA TensorRT
Looking for a C++ dev who knows TensorRT?
I'm looking for work. Hire me!
nvinfer1::Dims Class Reference

Structure to define the dimensions of a tensor. More...

Inheritance diagram for nvinfer1::Dims:
Collaboration diagram for nvinfer1::Dims:

Public Member Functions

 __attribute__ ((deprecated)) DimensionType type[MAX_DIMS]
 The type of each dimension, provided for backwards compatibility. More...
 

Public Attributes

int32_t nbDims
 The number of dimensions. More...
 
int32_t d [MAX_DIMS]
 The extent of each dimension. More...
 

Static Public Attributes

static const int32_t MAX_DIMS = 8
 The maximum number of dimensions supported for a tensor. More...
 

Detailed Description

Structure to define the dimensions of a tensor.

Note
: Currently the following formats are supported for layer inputs and outputs:
  • zero or more index dimensions followed by one channel and two spatial dimensions (e.g. CHW)
  • one time series dimension followed by one index dimension followed by one channel dimension (i.e. TNC)

TensorRT can also return an invalid dims structure. This structure is represented by nbDims == -1 and d[i] == 0 for all d.

TensorRT can also return an "unknown rank" dims structure. This structure is represented by nbDims == -1 and d[i] == -1 for all d.

Member Function Documentation

◆ __attribute__()

nvinfer1::Dims::__attribute__ ( (deprecated)  )

The type of each dimension, provided for backwards compatibility.

and will be removed in TensorRT 8.0.

Member Data Documentation

◆ MAX_DIMS

const int32_t nvinfer1::Dims::MAX_DIMS = 8
static

The maximum number of dimensions supported for a tensor.

◆ nbDims

int32_t nvinfer1::Dims::nbDims

The number of dimensions.

◆ d

int32_t nvinfer1::Dims::d[MAX_DIMS]

The extent of each dimension.


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