PyNEST implementation¶

Installing the python package `microcircuit`¶

The PyNEST implementation of the model is provided in the form of a python package microcircuit.

We recommend installing the python package inside a python environment:

Create a python environment
```
python -m venv venv
```
Activate the python environment:
```
source venv/bin/activate
```
Update pip:
```
pip install -U pip
```

The microcircuit python package can be installed by one of the following methods:

Install the package directly from PyPI:
```
pip install microcircuit
```
Optionally, you may want to install additional dependencies by using extra flags:

Option

Flag

runtime testing

test

development tools

dev

documentation build tools

doc

Example:
```
pip install microcircuit[dev,doc] 
```

Option	Flag
runtime testing	`test`
development tools	`dev`
documentation build tools	`doc`

Clone the repository and install the python package from the local sources:

 git clone https://github.com/inm-6/microcircuit-PD14-model
 pip install microcircuit-PD14-model/PyNEST

Install the package from the repository without cloning the entire repository:

pip install "git+https://github.com/INM-6/microcircuit-PD14-model.git/#egg=microcircuit&subdirectory=PyNEST"

Software requirements¶

NEST (see NEST installation)

We recommend installing NEST locally within the virtual environment:

git clone https://github.com/nest/nest-simulator
## for a specific <VERSION> (e.g. <VERSION>=v3.9), use
## git clone --depth 1 --branch <VERSION> https://github.com/nest/nest-simulator
cd nest-simulator
mkdir build
cd build
pip install -r ../requirements_pynest.txt
cmake ..
make
make install

Python 3.x
docopt-ng, matplotlib, numpy, psutil, ruamel.yaml, scipy (handled by python package dependencies)

Testing¶

To enable testing, make sure that the microcircuit package is installed with testing dependencies:

pip install microcircuit[test]

Executing

pytest

in the root directory of the repository microcircuit-PD14-model runs the test(s) in microcircuit-PD14-model/PyNEST/tests.

Usage¶

After installation, the microcircuit python package can be imported in a python application using

import microcircuit

See this example for a more detailed illustrations of how the package can be used.

Memory requirements¶

scaling factor (`= N_scaling = K_scaling`)	Memory
0.1 (default)	490 MB
0.2	1200 MB
0.5	4400 MB
1	14 GB

Performance benchmarking¶

Recent performance benchmarking results for the microcircuit model can be found here.

Implementation details¶

This implementation uses the iaf_psc_exp neuron and the static_synapse synapse models provided in [NEST]. The network is connected according to the fixed_total_number connection rule in NEST. The neuron dynamics is integrated in a time-driven manner using exact integration with a simulation step size sim_resolution (Rotter & Diesmann, 1999).

The PyNEST implementation runs with NEST 3.9 (Terhorst et al., 2025).

Simulation parameters (defaults)¶

Name	Value	Description
`sim_resolution`	0.1 ms	simulation time resolution (duration of one simulation step)
`t_presim`	500 ms	duration of pre-simulation phase (warm-up)
`t_sim`	1000 ms	duration of simulation phase
`rec_dev`	`spike_recorder`	recording device

References¶

Rotter & Diesmann (1999), Exact digital simulation of time-invariant linear systems with applications to neuronal modeling. Biological Cybernetics 81(5-6):381-402. doi:10.1007/s004220050570

Terhorst et al. (2025). NEST 3.9. Zenodo. doi:10.5281/zenodo.17036827

License¶

This project is licensed under GNU General Public License v2.0 or later. For details, see here.