torchosr -- a PyTorch extension package for Open Set Recognition models evaluation in Python
Joanna Komorniczak, Pawel Ksieniewicz
TL;DR
The paper introduces torchosr, a PyTorch-compatible package designed to standardize and simplify Open Set Recognition experimentation. It provides two state-of-the-art OSR methods (Thresholded Softmax and Openmax), a comprehensive data handling stack (Base Datasets, Data, utilities), and tools to generate diverse problem configurations across Openness with cross-validation. By offering an end-to-end workflow—from dataset loading to multi-configuration evaluation and visualization—torchosr enables reproducible benchmarking and fair comparisons against reference implementations. The work addresses a key gap in OSR research for accessible, open-source, and well-documented experimentation infrastructure, with potential to accelerate method development and reproducibility in the field.
Abstract
The article presents the torchosr package - a Python package compatible with PyTorch library - offering tools and methods dedicated to Open Set Recognition in Deep Neural Networks. The package offers two state-of-the-art methods in the field, a set of functions for handling base sets and generation of derived sets for the Open Set Recognition task (where some classes are considered unknown and used only in the testing process) and additional tools to handle datasets and methods. The main goal of the package proposal is to simplify and promote the correct experimental evaluation, where experiments are carried out on a large number of derivative sets with various Openness and class-to-category assignments. The authors hope that state-of-the-art methods available in the package will become a source of a correct and open-source implementation of the relevant solutions in the domain.