neuro_op

Package to research information establishment in networks of topology-restricted active inference nodes.

Installation

From source

Clone this repository via HTTPS

$ git clone https://github.com/Priesemann-Group/neuro_op.git

or SSH (requires SSH key setup)

$ git clone [email protected]:Priesemann-Group/neuro_op.git

Subsequently, change into the just-cloned directory and, if you have not yet set up a conda environment with needed packages, do so:

$ cd neuro_op
$ # If not yet existing, create conda environment with needed packages. This also installs pip within that environment
$ conda env create -f environment.yml

To install the package within this environment (named neuro_op), activate the package and install from the repo's directory

$ conda activate neuro_op
$ pip install -e .	    # This assumes you have been 'cd'ing into the neuro_op directory

This will install the project in editable mode, meaning that changes in the local source code will apply without requiring a reinstall of the package.

To access this environment via Jupyter (e.g., via JupyterLab), install an ipykernel for it:

$ python -m ipykernel install --user --name neuro_op --display-name "Python 3.11 (neuro_op)"