A set of named ONNX tensors for multi-input model inference. More...

#include <onnx_bridge.hpp>

Public Member Functions
bool	is_valid () const
	Check whether all tensors are valid and the set is non-empty.

Public Attributes
std::vector< std::string >	names
	Model input names (parallel with tensors)

std::vector< OnnxTensorInfo >	tensors
	Prepared tensor infos (parallel with names)

Detailed Description

A set of named ONNX tensors for multi-input model inference.

Names and tensors are stored in parallel vectors. The names correspond to model input names as defined in the ONNX model graph, and the tensors hold OnnxTensorInfo structs ready for OrtApi::CreateTensorWithDataAsOrtValue.

See also: prepare_inputs_for_onnx

Definition at line 235 of file onnx_bridge.hpp.

Member Function Documentation

◆ is_valid()

bool signet::forge::OnnxInputSet::is_valid ( ) const

inline

Check whether all tensors are valid and the set is non-empty.

Returns: true if names and tensors have the same non-zero size and every OnnxTensorInfo passes is_valid().

Definition at line 243 of file onnx_bridge.hpp.

Member Data Documentation

◆ names

std::vector<std::string> signet::forge::OnnxInputSet::names

Model input names (parallel with tensors)

Definition at line 236 of file onnx_bridge.hpp.

◆ tensors

std::vector<OnnxTensorInfo> signet::forge::OnnxInputSet::tensors

Prepared tensor infos (parallel with names)

Definition at line 237 of file onnx_bridge.hpp.

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

include/signet/interop/onnx_bridge.hpp