Freedom251 › Statistical Analysis Software

MBPLS

Free plan Linux

What MBPLS is

MBPLS is a free Python package for Partial Least Squares prediction using univariate or multivariate outcomes from multiple data blocks. It provides KERNEL, NIPALS, and UNIPALS algorithms for multiblock modeling, together with SIMPLS for fast predictions. The MBPLS class supports fitting, prediction, data transformation, and plotting model values for exploratory analysis. Documentation describes support for missing data and algorithms optimized for large data matrices. Its interface is described as compatible with the Scikit-learn API and model-selection toolbox. NIPALS is the documented choice for sparse data, and sparse_data must be set to True. Python 3 installation uses `pip install mbpls`; dependencies including scikit-learn, numpy, and scipy may require upgrades. The package can measure each X block’s contribution to predicting Y. Examples are provided as real-world Jupyter Notebooks. It is BSD-3-Clause licensed and lists bioinformatics, chemometrics, data fusion, metabolomics, and machine learning among its topics.

Who it suits

MBPLS suits Python users working with multiblock prediction, including bioinformatics, chemometrics, data fusion, metabolomics, and machine learning. It is intended for users needing univariate or multivariate outcomes and Scikit-learn-style workflows.

What it does well

Free Python package
Supports univariate and multivariate outcomes
Offers several multiblock algorithms
Designed to handle missing data
Includes plotting and block-contribution measures

What to know first

Linux platform listed
Sparse data requires NIPALS
Dependencies may require upgrades

Verdict

MBPLS combines several multiblock PLS algorithms with fitting, prediction, transformation, and plotting. Its Linux platform and dependency requirements should be checked before installation.

Facts from MBPLS's own pages

Algorithms: The package provides KERNEL, NIPALS, and UNIPALS for multiblock modeling, as well as SIMPLS for fast predictions. · mbpls.readthedocs.io, 12 Oct 2026
Block contribution: The introduction example says MB-PLS provides a measure of how much each X block contributes to predicting Y. · mbpls.readthedocs.io, 12 Oct 2026
Dependencies: The installation instructions say some dependencies may require upgrades, naming scikit-learn, numpy, and scipy. · mbpls.readthedocs.io, 12 Oct 2026
Examples: The documentation links to real-world Jupyter Notebook examples in the project's GitHub repository. · mbpls.readthedocs.io, 12 Oct 2026
Free plan: Yes · mbpls.readthedocs.io, 24 Sept 2026
Installation: The documentation gives `pip install mbpls` as the Python 3 installation command. · mbpls.readthedocs.io, 12 Oct 2026
Intended fields: The repository lists bioinformatics, chemometrics, data fusion, metabolomics, and machine learning among its topics. · github.com, 12 Oct 2026
Large matrices: The documentation says its algorithms are optimized for robust performance on large data matrices. · mbpls.readthedocs.io, 12 Oct 2026
License: The project repository identifies its license as BSD-3-Clause. · github.com, 12 Oct 2026
Missing data: The package documentation says it is designed to handle missing data. · mbpls.readthedocs.io, 12 Oct 2026
Model methods: The MBPLS class supports fitting, prediction, transforming data, and plotting model values. · mbpls.readthedocs.io, 12 Oct 2026
Multiple data blocks: MBPLS predicts a response vector or matrix from multiple data blocks. · mbpls.readthedocs.io, 12 Oct 2026
Outcomes: It supports prediction of univariate or multivariate outcomes. · mbpls.readthedocs.io, 12 Oct 2026
Plotting: The quick-start example uses a built-in plot method for exploratory analysis of multiblock PLS models. · mbpls.readthedocs.io, 12 Oct 2026
Purpose: MBPLS is a Python package for Partial Least Squares prediction modeling with univariate or multivariate outcomes. · mbpls.readthedocs.io, 12 Oct 2026
Regression models: Yes · mbpls.readthedocs.io, 24 Sept 2026
Scikit-learn API: The package uses an interface described as compatible with the commonly known Scikit-learn API and its model selection toolbox. · mbpls.readthedocs.io, 12 Oct 2026
Scikit-learn interface: The package uses a commonly known Scikit-learn API and its model selection toolbox. · mbpls.readthedocs.io, 12 Oct 2026
Scripting support: Yes · mbpls.readthedocs.io, 24 Sept 2026
Sparse data limit: NIPALS is the only documented algorithm that handles sparse data, and sparse_data must be set to True to allow it. · mbpls.readthedocs.io, 12 Oct 2026

Where MBPLS stands on Freedom251

Statistical Analysis Software#67 of 106