reusex::vision::libtorch::LibTorchYolo Class Reference

YOLO instance segmentation model using LibTorch (TorchScript). More...

#include <Yolo.hpp>

Inheritance diagram for reusex::vision::libtorch::LibTorchYolo:

Collaboration diagram for reusex::vision::libtorch::LibTorchYolo:

Public Member Functions
	LibTorchYolo (const std::filesystem::path &model_path, bool use_cuda=false)
	Construct and load a YOLO model from a TorchScript file.
std::vector< IDataset::Pair >	forward (const std::span< IDataset::Pair > &input) override
	Run YOLO inference on a batch of images.
Public Member Functions inherited from reusex::vision::IModel
virtual	~IModel ()=default

Static Public Member Functions
static std::unique_ptr< LibTorchYolo >	create (const std::filesystem::path &model_path, bool use_cuda=false)
	Factory method for creating a LibTorchYolo instance.
Static Public Member Functions inherited from reusex::vision::IModel
static std::unique_ptr< IModel >	create (const std::filesystem::path &model_path, bool use_gpu=false)

Detailed Description

YOLO instance segmentation model using LibTorch (TorchScript).

Loads a TorchScript YOLO segmentation model and implements the IModel interface for use with the backend-agnostic annotation pipeline.

Definition at line 24 of file Yolo.hpp.

Constructor & Destructor Documentation

LibTorchYolo()

reusex::vision::libtorch::LibTorchYolo::LibTorchYolo ( const std::filesystem::path & model_path, bool use_cuda = false )

explicit

Construct and load a YOLO model from a TorchScript file.

Parameters

model_path	Path to the .pt TorchScript model file.
use_cuda	Whether to use CUDA for inference (falls back to CPU).

Member Function Documentation

create()

std::unique_ptr< LibTorchYolo > reusex::vision::libtorch::LibTorchYolo::create ( const std::filesystem::path & model_path, bool use_cuda = false )

static

Factory method for creating a LibTorchYolo instance.

Parameters

model_path	Path to the .pt TorchScript model file.
use_cuda	Whether to use CUDA for inference (defaults to false).

Returns

Unique pointer to the created model.

forward()

std::vector< IDataset::Pair > reusex::vision::libtorch::LibTorchYolo::forward ( const std::span< IDataset::Pair > & input )

overridevirtual

Run YOLO inference on a batch of images.

Each input Pair must contain a LibTorchData with a letterboxed image. Output Pairs contain the same LibTorchData with label_image populated.

Parameters

input

Span of (data, index) pairs from the dataset.

Returns

Vector of (data, index) pairs with label_image set.

Implements reusex::vision::IModel.

---

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

• libs/reusex/include/vision/libtorch/Yolo.hpp