ANNarchy 5.0.3

Artificial Neural Networks architect

ANNarchy

ANNarchy (Artificial Neural Networks architect) is a parallel and hybrid simulator for distributed rate-coded or spiking neural networks. The core of the library is written in C++ and distributed using openMP or CUDA. It provides an interface in Python for the definition of the networks. It is released under the GNU GPL v2 or later.

• Source code: github.com/ANNarchy/ANNarchy
• Documentation: annarchy.github.io
• Forum: google forum
• Bug reports and feature requests: Issue Tracker.

Citation

If you use ANNarchy for your research, we would appreciate if you cite the following paper:

Vitay J, Dinkelbach HÜ and Hamker FH (2015). ANNarchy: a code generation approach to neural simulations on parallel hardware. Frontiers in Neuroinformatics 9:19. doi:10.3389/fninf.2015.00019

Authors

• Julien Vitay ([e-mail]).
• Helge Ülo Dinkelbach ([e-mail]).
• Oliver Maith ([e-mail])
• Fred Hamker ([e-mail]).

Installation

Using pip, you can install the latest stable release:

pip install ANNarchy

See https://annarchy.github.io/Installation for further instructions.

Platforms

• GNU/Linux
• MacOS X
• Windows (inside WSL2)

Dependencies

• python >= 3.10 (with the development files, e.g. python-dev or python-devel)
• g++ >= 7.4 or clang++ >= 3.4
• cmake >= 3.17
• setuptools >= 65.0
• nanobind >= 2.4.0
• cython >= 3.0
• numpy >= 1.21
• sympy >= 1.11
• scipy >= 1.9
• matplotlib >= 3.0
• tqdm >= 4.60

Recommended:

• lxml (to save the networks in .xml format).
• h5py (to export data in .h5 format).
• pandoc (for report()).
• tensorflow (for the ann_to_snn_conversion extension)
• tensorboardX (for the logging extension).