Welcome to SemiPy's documentation

Warning

SemiPy is not released yet. Stay tuned, the repository should be public soon at https://github.com/SemiPy.

A PyTorch based Python library dedicated to Semi-Supervised Learning (SSL).

Introduction

Welcome to the documentation of SemiPy, an open-source Python library designed specifically for Semi-Supervised Learning (SSL) using the power of PyTorch. The goal of SemiPy is to be a toolbox designed to tackle SSL experiments and real-world problematics. It includes different SSL methods, datasets and useful tools for SSL. With SemipY, we hope that you will be able to unlock the potential of SSL in your machine learning project.

See Getting Started for instructions on how to use the library and diverse tutorials.

See About for more informations about the team behind SemiPy and the used licence.

What is Semi-Supervised Learning ?

Semi-Supervised Learning (SSL) is a field of machine learning that addresses the challenges posed by limited labelled data by leveraging the potential hidden information of unlabelled data. SSL has emerged as a powerful machine learning strategy, bridging the gap between the efficiency of unsupervised learning and the performance of supervised learning. By using the unlabelled data, the model can learn from the underlying structure of the data distribution, even when explicit labels are unavailable.