Description:

Nimfa is an open-source Python library that provides a unified interface to nonnegative matrix factorization algorithms. It includes implementations of state-of-the-art factorization methods, initialization approaches, and quality scoring. Both dense and sparse matrix representation are supported.

Matrix Factorization Methods

• BD - Bayesian nonnegative matrix factorization Gibbs sampler
• BMF - Binary matrix factorization
• ICM - Iterated conditional modes nonnegative matrix factorization
• LFNMF - Fisher nonnegative matrix factorization for learning local features
• LSNMF - Alternating nonnegative least squares matrix factorization using projected gradient method for subproblems
• NMF - Standard nonnegative matrix factorization with Euclidean / Kullback-Leibler update equations and Frobenius / divergence / connectivity cost functions
• NSNMF - Nonsmooth nonnegative matrix factorization
• PMF - Probabilistic nonnegative matrix factorization
• PSMF - Probabilistic sparse matrix factorization
• SNMF - Sparse nonnegative matrix factorization based on alternating nonnegativity constrained least squares
• SNMNMF - Sparse network-regularized multiple nonnegative matrix factorization

Initialization Methods

• Random
• Fixed
• NNDSVD
• Random C
• Random VCol

Quality Measures

• Distance
• Residuals
• Connectivity matrix
• Consensus matrix
• Entropy of the fitted NMF model
• Dominant basis components computation
• Explained variance
• Feature score computation representing its specificity to basis vectors
• Computation of most basis specific features for basis vectors
• Purity
• Residual sum of squares (rank estimation)
• Sparseness
• Cophenetic correlation coefficient of consensus matrix (rank estimation)
• Dispersion
• Factorization rank estimation
• Selected matrix factorization method specific

Utils

• Fitted factorization model tracker across multiple runs
• Residuals tracker across multiple factorizations / runs

Changes to previous version:

Initial Announcement on mloss.org.

BibTeX Entry: Download

Supported Operating Systems: Agnostic

Data Formats: Agnostic

Tags: Nonnegative Matrix Factorization, Initialization Methods, Quality Measures

Archive: download here

Comments

No one has posted any comments yet. Perhaps you'd like to be the first?

Leave a comment

You must be logged in to post comments.