Skip to content

SpikeShip: A method for fast, unsupervised discovery of high-dimensional neural spiking patterns.

License

Notifications You must be signed in to change notification settings

bsotomayorg/SpikeShip

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

26 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

SpikeShip: A method for fast, unsupervised discovery of high-dimensional neural spiking patterns.

The Python 3 module spikeship which implements the fast, unsupervised dissimilarity measure described in Sotomayor-Gómez, B., L., Battaglia, F. and Vinck, M. (2023). "SpikeShip: A method for fast, unsupervised discovery of high-dimensional neural spiking patterns". Article, PLoS CB.

Setup

The dependencies can be installed by running ./env_setup.sh <ENV_NAME> with the optional argument specifying the target environment (which must be source-able). To setup the module, run python setup.py install. A jupyter notebook is available in notebooks/, along with a demo dataset, showing an example workflow for the SpikeShip methods and its comparison with SPOTDis.

Linux (Anaconda)

  1. conda create -n spikeship python=3
  2. ./env_setup.sh spikeship
  3. source activate spikeship
  4. python setup.py install

Windows (Anaconda)

  1. conda create -n spikeship python=3.6.5
  2. conda activate spikeship
  3. conda install numba
  4. conda install ipykernel
  5. conda install matplotlib
  6. conda install scikit-learn
  7. conda install joblib
  8. python setup.py install

Further notes

The software requirements/dependencies are the same from the work SPOTDis, the implementation of the Spike Pattern Optimal Transport Dissimilarity described in Grossberger, L., Battaglia, F. and Vinck, M. (2018). Unsupervised clustering of temporal patterns in high-dimensional neuronal ensembles using a novel dissimilarity measure. PLoS CB.