A set of python modules for fitting, testing, and visualizing parameters of neural encoding models (NEMs).
Neural encoding models are models that try and predict neural activity given a stimulus. For example, we can fit models to predict the spiking activity of neurons in the retina or V1 in response to a visual stimulus displayed on a computer monitor.
We include general tools that allow you to fit the parameters of encoding models of any functional form. Additionally, we provide specific classes to fit linear-nonlinear (LN) and cascaded (2-layer) linear-nonlinear (LN-LN) models to data.
Used in the paper: Inferring hidden structure in multilayered neural circuits.
git clone git@github.com:ganguli-lab/nems.git
cd nems
pip install -r requirements.txt
python setup.py installNumpy, scipy, pandas and the proxalgs package.
Pull requests welcome! Please stick to the NumPy/SciPy documentation standards
We use sphinx for documentation and nose for testing.