AxoNN is a parallel framework for training deep neural networks.

Prior to the installation, PyTorch must already be installed.

pip install axonn

AxoNN is an open source project. We welcome contributions via pull requests, and questions, feature requests, or bug reports via issues.

If you are referencing AxoNN in a publication, please cite the following paper:

• Siddharth Singh, Abhinav Bhatele. AxoNN: An asynchronous, message-driven parallel framework for extreme-scale deep learning In Proceedings of the IEEE International Parallel & Distributed Processing Symposium (IPDPS '22). IEEE Computer Society, May 2022.

AxoNN is distributed under the terms of the Apache License (Version 2.0) with LLVM Exceptions.

All contributions must be made under the Apache License (Version 2.0) with LLVM Exceptions.

See LICENSE for details.

SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception