pip install pyPLNmodelsHome page 🏠
The Poisson lognormal model and its variants are used for the analysis of multivariate count data. This package implements efficient algorithms to extract meaningful insights from complex and difficult-to-interpret multivariate count data. It is designed to scale on large datasets, although it has memory limitations.
1 In-depth pyPLNmodels tutorials
Unlike the getting started notebook, and the quickstart, this is an in-depth tutorial, covering in detail different parts.
1.1 Installation
First of all, you should run:
1.2 Tutorials
- Check detailed formulas of each model.
- Check how to specify a model.
- For a basic analysis of high-dimensional count data, see this tutorial.
- If your data is zero-inflated, you may consider the zero-inflation tutorial.
- If your data has temporality or one-dimensional spatiality, you should consider the time-series tutorial for count data.
- If you want to cluster your samples, be it in a supervised or unsupervised way, you should consider the clustering tutorial for count data.
- You may do a network analysis, that is, see the link between variables, through the network analysis tutorial for count data.
References
Batardière, Bastien, Joon Kwon, and Julien Chiquet. 2024. “pyPLNmodels: A Python Package to Analyze Multivariate High-Dimensional Count Data.” Journal of Open Source Software.