Repository files navigation

A neural network library in c++ aimed at providing a simple, modularized framework for deep learning that is accelerated for heterogeneous architectures. MagmaDNN's releases are located athttps://bitbucket.org/icl/magmadnn (andhere), while active development occurs athttps://github.com/MagmaDNN/magmadnn/tree/dev. If you're looking to contribute or submit a pull-requests/issues, then please do so on the github development repository.

In version 1.0 MagmaDNN offers a strong tensor core with standard machine learning and DNN functionalities built around it. For nightly development builds use the github repository linked above.

MagmaDNN is optimized towards heterogeneous architectures (multi-core CPU and GPU), so it is advised to use with a modern NVIDIA GPU. However, MagmaDNN does support a CPU only install. This is mainly meant for testing and is not nearly as optimized as the GPU version.

The documentation can be found on thedocs site. For the most recent version of the documentation see thebuild & install tutorial on how to build the docs from source. Thetodo page contains information on the future of the package and thetroubleshooting page walks through common issues and there solution.

There are several tutorials indocs/tutorials. These give an introduction into installing and using the library.

For examples of what MagmaDNN code looks like see theexamples/ folder. If MagmaDNN is downloaded and installed, then the examples can be made and run withmake examples.

All development takes place on thegithub site.

author: Daniel Nichols

co-authors: Florent Lopez, Sedrick Keh, Rocco Febbo