Skip to content

Welcome to CorrelatedCounts

Getting Started

In order to run the correlated count models, you will need to install Python to your computer. It is most helpful to use miniconda, and then create an environment that will be used specifically with the correlated count package. The dependencies for this package are numpy and scipy.

Installation

To install the package, clone the GitHub repository here:

git clone https://github.com/mbannick/CorrelatedCounts.git

Once you have cloned the package, activate your conda envirionment and run

cd CorrelatedCounts
python setup.py install

The package name is ccount.

We recommend using Jupyter Notebooks (documentation here) in your environment, which you can install with

pip install jupyter

and then running the command

jupyter-notebook

Once inside the notebook, you can import all of the functions that are described in the code documentation, including model specification, fitting, and prediction.