A port of the code from Chenkov et al. to current versions of Brian2 pipeline, as a stsarting point for current investigations
Currently targeting:
- brian2=2.2.2.1
- numpy=1.18.1
- python=3.6.10
create brian2 conda environment with
conda env create -f environment.yml
conda activate brian2
From a Mac you can get the literal configuration I'm on by using environment_mac.yml instead.
You can then install the package probably with
python setup.py develop
This code reproduces some results from the paper:
Memory replay in balanced recurrent networks. N. Chenkov, H. Sprekeler, R. Kempter PLoS Comput. Biol., 13(1):e1005359, 2017. DOI: 10.1371/journal.pcbi.1005359
The code managed to cold-rot, and old packages need to be imported. One way to run the program is to create a virtual environment. The following lines create a virtual environment in python2, activate it, and install old versions of scipy, numpy and brian through pip:
virtualenv2 assembly_vm
source assembly_vm/bin/activate
pip install scipy==0.17
pip install numpy==1.10.4
pip install brian=1.4.1
The main file is assemblyseq.py. 'run assemblyseq' in a python shell should illustrate activity propagation through an assembly sequence.