Deep learning for biological sequences with fastai
Project description
Welcome to fastBio
fastBio is a library for manipulating data and creating and training deep learning models for biological sequencing data. It is an extension of the fastai v1 library.
A number of pretrained models for biological sequencing data - a universal language model as well as several downstream transfer learning tasks - can be loaded with the fastBio LookingGlass and LookingGlassClassifier classes. These models are available for download at the sister repository LookingGlass: https://github.com/ahoarfrost/LookingGlass
If you find fastBio or LookingGlass useful, please cite the preprint:
Hoarfrost, A., Aptekmann, A., Farfanuk, G. & Bromberg, Y. Shedding Light on Microbial Dark Matter with A Universal Language of Life. bioRxiv (2020). doi:10.1101/2020.12.23.424215. https://www.biorxiv.org/content/10.1101/2020.12.23.424215v2
Installation
You can install fastBio with pip (python 3 only):
pip3 install fastBio
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.