Super-Selfish: Self-Supervised Learning on Images with PyTorch

TL;DR

Super-Selfish is a user-friendly PyTorch framework that simplifies self-supervised learning on images, supporting 13 algorithms and enabling quick pretraining of neural networks with minimal code.

Contribution

It introduces an easy-to-use, modular PyTorch framework supporting multiple self-supervised algorithms for image learning, facilitating rapid pretraining with minimal effort.

Findings

Supports 13 self-supervised algorithms

Enables pretraining with two lines of code

Flexible and modular design

Abstract

Super-Selfish is an easy to use PyTorch framework for image-based self-supervised learning. Features can be learned with 13 algorithms that span from simple classification to more complex state of theart contrastive pretext tasks. The framework is easy to use and allows for pretraining any PyTorch neural network with only two lines of code. Simultaneously, full flexibility is maintained through modular design choices. The code can be found at https://github.com/MECLabTUDA/Super_Selfish and installed using pip install super-selfish.