Dragon is a C(Computation)G(Graph)V(Virtual)M(Machine) based distributed deep learning framework.

Our goal is to reduce the unnecessary structures or interfaces. Therefore, in addition to feed or fetch, the last thing is designing a objective function through all available operators.

Besides, we demonstrate that a cross-frameworks frontend(Deep Learning VirtualBox) is feasible, and further more, will get benefit from all participating crucial interfaces especially when one is not reasonable.

Dragon 0.2.1 Released - The preliminary documentation, and massive known bugs are fixed.

Dragon is released under the BSD 2-Clause license.

Please cite Dragon in your publications if it helps your research:

@article{pan2017dragon,
  Author = {Pan, Ting},
  Journal = {arXiv preprint arXiv:1707.08265},
  Title = {Dragon: A Computation Graph Virtual Machine Based Deep Learning Framework},
  Year = {2017}

}