lucidrains/byol-pytorch
A PyTorch implementation of BYOL, a self-supervised learning technique for training neural networks on unlabeled image data.
Not currently ranked — collecting fresh signals.
star history
Implements the BYOL self-supervised learning algorithm from DeepMind in PyTorch. The library provides a wrapper module that can be applied to any image-based neural network to enable training on unlabeled data through prediction of latent representations. It eliminates the need for labeled data or contrastive learning approaches, achieving competitive results with SimCLR using a simpler methodology.
Frequently asked
- What is lucidrains/byol-pytorch?
- A PyTorch implementation of BYOL, a self-supervised learning technique for training neural networks on unlabeled image data.
- Is byol-pytorch open source?
- Yes — lucidrains/byol-pytorch is open source, released under the MIT license.
- What language is byol-pytorch written in?
- lucidrains/byol-pytorch is primarily written in Python.
- How popular is byol-pytorch?
- lucidrains/byol-pytorch has 1.9k stars on GitHub.
- Where can I find byol-pytorch?
- lucidrains/byol-pytorch is on GitHub at https://github.com/lucidrains/byol-pytorch.