Installation#
Dendrify is included in the Python package index: https://pypi.org/project/dendrify.
The easiest way to install it is through pip, using the command:
pip install dendrify
The above command will automatically install Brian 2.5.4, if it is not already installed. If you wish to work with a different Brian version, you can install Dendrify without its dependencies using the command:
pip install dendrify --no-deps
and then install the desired version of Brian separately (see bellow).
Dependencies#
Brian 2 (required) is a simulator for spiking neural networks. It is written in Python and is available on almost all platforms. Brian is designed to be easy to learn and use, highly flexible and easily extensible.
Networkx (optional) is a Python package for the creation, manipulation, and study of the structure, dynamics, and functions of complex networks. If you wish Dendrify to have access to some experimental model visualization features, you can install it using the command:
pip install networkx
GPU support#
Dendrify is compatible with Brian2CUDA, a Python package for simulating spiking neural networks on graphics processing units (GPUs). Brian2CUDA is an extension of Brian2 that uses the code generation system of the latter to generate simulation code in C++/CUDA, which is then executed on NVIDIA GPUs.