Neuroptimus is an open-source framework for solving parameter optimization problems, with many additional features (including a GUI) to support typical use cases in neuroscience. The neuroptimus/new_test_files folder in the repository contains several examples of using Neuroptimus, with detailed PDF guides to setting up and running these use cases in the Neuroptimus GUI.

Installation

Get a copy of Neuroptimus

Install git and type:

git clone https://github.com/KaliLab/neuroptimus.git

Dependencies

The following python libraries are required:

• python
• numpy
• eFEL
• matplotlib

The following libraries are recommended:

• neuron
• scipy
• PyQt5
• inspyred
• pyelectro
• Pygmo
• bluepyopt
• ipyparallel
• nest

You can get required or recommended libraries with the following command:

pip install -r requirements_minimal.txt

or

pip install -r requirements_full.txt

You can get individual package with pip:

pip install neuron

Build documentation

If you should require a local copy of the Neuroptimus documentation, you need a working install of Sphinx, then run the command:

sphinx-build ./doc <local build directory>

from the top-level neuroptimus directory where should be replaced with a custom filepath.

Test Platforms

The package was tested on the following systems:

1. Ubuntu 22.04.1 LTS
2. Ubuntu 20.04.6 LTS 
3. Fedora release 32 (Thirty Two) (neurofedora)

Running Neuroptimus

You can run Neuroptimus (with a GUI) directly from its installation folder with:

python neuroptimus.py -g

Or for the command line version (you must specify a configuration file as well):

python neuroptimus.py -c example.xml

Developers

Project leader:

    Szabolcs Káli

        [email protected]

Lead developer:

    Máté Mohácsi

        [email protected]

Past developers:

    Péter Friedrich

    Sára Sáray

    Márk Patrik Török