Swift for TensorFlow: A portable, flexible platform for deep learning

Part of Proceedings of Machine Learning and Systems 3 (MLSys 2021)

Bibtex Paper

Authors

Brennan Saeta, Denys Shabalin

Abstract

Swift for TensorFlow is a deep learning platform that scales from mobile devices to clusters of hardware accelerators in data centers. It combines a language-integrated automatic differentiation system and multiple Tensor implementations within a modern ahead-of-time compiled language oriented around mutable value semantics. The resulting platform has been validated through use in over 30 deep learning models and and has been employed across data center and mobile applications.