Skip to content

timwroge/LFADS_CNBC_Pipeline

Repository files navigation

LFADS CNBC Pipline


In order to interface with the LFADS software, a matlab class must be created that takes a input data format (such as .mat) and outputs a instance of firing rates and other data from the trial in order to run the software and create the latent factor analysis instance.

Example: Lorenz Dataset

The LFADS software includes an example synthetic dataset and an example class that reads that dataset.

The code for the matlab instance for the batista lab is given based off this program.

Using this pipeline

Just clone this repository into some place like a directory called +CNBCPipeline or for a specific experiment, +NelsonHandExperiment. From there, the run method should be modified to referenece the experiment's specific datasets.

Install

Just run ./install on a linux machine. After installation make sure to source the bash profile by running: source ~/.bashrc

About

The LFADS pipeline for the batista lab at the CNBC

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published