Brainstorm makes working with neural networks fast, flexible and fun.
Combining lessons from previous projects with new design elements, and written entirely in Python, Brainstorm has been designed to work on multiple platforms with multiple computing backends.
If you have the pre-requisite packages python-dev libhdf5-dev libopenblas-dev, quick installation is simply: pip install brainstorm.
If you have CUDA set up already, pip install brainstorm[all] will install all optional dependencies.
See Installation instructions in the documentation for information about pre-requisites, GPU support, and various platforms.
Brainstorm is under active development and is currently in beta.
Documentation is in progress, and hosted on ReadTheDocs.
A good point to start is the brief walkthrough of the cifar10_cnn.py example.
If you wish, you can also run the data preparation scripts (data directory) and look at some basic examples (examples directory).
If you have any suggestions or questions, please post to the Google group.
If you encounter any errors or problems, please let us know by opening an issue.
MIT License. Please see the LICENSE file.
Klaus Greff and Rupesh Srivastava would like to thank Jürgen Schmidhuber for his continuous supervision and encouragement. Funding from EU projects NASCENCE (FP7-ICT-317662) and WAY (FP7-ICT-288551) was instrumental during the development of this project. We also thank Nvidia Corporation for their donation of GPUs.