Welcome to Speos’s documentation!#
Speos is a framework built on PyG (PyTorch Geometric) and PyTorch to easily train and evaluate Graph Neural Networks (GNNs) and other machine learning models for gene classification.
It handles the integration of various types of genomic data, as well as the complete training and evaluation process. For the motivation behind the framework you can check our preprint with exciting applications and showcases.
Speos is designed as a low code platform, meaning that users will have to write little to no code to access the full bandwidth of state-of-the-art methods and datasets. It is possible to extend Speos to your needs, which might come with a minimum of implementation effort, as detailed in the later chapters of this documentation.
For now, you can go through the chapters of this documenation one by one and learn how to install, use, adapt and extend Speos to power your experiments!
Note
If you are facing any issues or have questions, do not hesitate to file a bug report or feature request via GitHub Issues.