Skip to content
catubc edited this page Feb 11, 2021 · 28 revisions

Introduction

Yass is an open source, automated spike sorting software developed as a platform for sorting high-firing rate, high-collision, large-spatial extent spikes. The YASS manuscript is available on Biorxiv (and is currently in submission in a peer-reviewed journal): https://www.biorxiv.org/.

Installation

YASS is available on AWS services as an Amazon Machine Image (AMI) on EC2 instances (see Installation AWS).

YASS can also be installed and run on a local workstation (see Installation Local). We have tested installation on systems running Ubuntu 18.04 with a minimum of 16GB of CPU-RAM and a GPU with at least 6GB of GPU-RAM, but other operating systems should also be supported. We also recommend using solid-state-drives (SSDs) with at least 5-6 times of free space compared to the size of the dataset.

Config file parameters

There are several user configurable parameters in the config.yaml file that must be changed for each run. Please review the Data Formats and Configuration Wiki for more details.

Running YASS

YASS can be run from the command line using the command:

yass sort config.yaml

It can also be run from the GUI using the RUN Yass button.

GUI

If YASS is installed on a local machine it can be used via a basic GUI (GUI) via the command:

yass gui

Retraining Neural Networks

The default YASS NNs can be retrained using an automated interface. For more information please see Retraining NNs.