mvlearn is a Python module for multiview learning.
In many data sets, there are multiple measurement modalities of the same subject, i.e. multiple X matrices (views) for the same class label vector y. For example, a set of diseased and healthy patients in a neuroimaging study may undergo both CT and MRI scans. Traditional methods for inference and analysis are often poorly suited to account for multiple views of the same subject as they cannot account for complementing views that hold different statistical properties. While single-view methods are consolidated in well-documented packages such as scikit-learn, there is no equivalent for multiview methods. In this package, we provide a well-documented and tested collection of utilities and algorithms designed for the processing and analysis of multiview data sets.
Python is a powerful programming language that allows concise expressions of network algorithms. Python has a vibrant and growing ecosystem of packages that mvlearn uses to provide more features such as numerical linear algebra. In order to make the most out of mvlearn you will want to know how to write basic programs in Python. Among the many guides to Python, we recommend the Python documentation.
Currently, mvlearn is supported for Python 3.6, 3.7, and 3.8.
mvlearn was developed during the end of 2019 by Richard Guo, Ronan Perry, Gavin Mischler, Theo Lee, Alexander Chang, Arman Koul, and Cameron Franz, a team out of the Johns Hopkins University NeuroData group.
mvlearn is a Python package of multiview learning tools.