Code for Trained recurrent neural networks develop phase-locked limit cycles in a working memory task.
First install the conda environment:
conda env create -f phase_env_ubu.yml, if you are on Ubuntu, or
conda env create -f phase_env_mac.yml if you are on MacOS, then
activate it.
You can train an RNN models by running python rnn_scripts/run_training.py
All the paper figures can be recreated with the notebooks inside the generate_figures folder
To create the 3D plots you need to allow the Mayavi plug-in:
$ jupyter nbextension install --py mayavi --user
$ jupyter nbextension enable --py mayavi --user
Paper written by: Matthijs Pals, Jakob Macke and Omri Barak, code written by: Matthijs Pals