The only requirement is having Docker installed on your machine.
If you want to use it without Docker, list of the required python packages that the code uses are in requirements.txt
This is the quickest way to get this code to run. You can readily use this package from it's docker image. To download and run the CSsorter docker, use the following command:
docker run -it -p 8888:8888 --name cs_sorter -v /path/to/local/dir:/run/dmount:z kkarbasi/cssorter
After running the above command, you can access a Jupyter notebook with sample run code (get_started) from your browser at:
http://127.0.0.1:8888
The /run/dmount path inside the Docker container will be mounted to /path/to/local/dir on your local machine
For more information on using docker containers see here
clone the repository: git clone https://github.com/kkarbasi/cssorter.git
Installing the packages:
pip3 install --no-cache-dir -r requirements.txt
Then follow the instructions on the get_started Jupyter notebook to run it on the provided sample data or your own data.
If you want to use it in your own Python script:
Assuming the voltage signal is loaded in a numpy vector named voltage and the sampling frequency is in Fs
from cssorter.spikesorter import ComplexSpikeSorter
css = ComplexSpikeSorter(voltage, 1.0/Fs)
css.run()
css.num_gmm_components: default=5; for noisy data use larger numbers
css.cs_num_gmm_components: default=5;
The voltage window in which we check for CS signature. If the cell is very high frequeny, use shorter post_window
css.pre_window: default=0.0002 seconds;css.post_window: default=0.005 seconds;
css.run() arguments:
use_filtered = (True/False): If set to True, spike detection is performed on filtered(Sav-Golay) data.spike_detection_dir = ('min'/'max'): The direction towards which the adaptive spike detector should look for spikes.align_spikes_to = ('min'/'max'): If set to 'min' ('max') spikes are aligned to their minimum (maximum).remove_overlaps = (True/False)