Package: nmfkc 1.0.2

nmfkc: Non-Negative Matrix Factorization with Kernel Covariates

Performs Non-negative Matrix Factorization (NMF) with Kernel Covariates. Given an observation matrix and kernel covariates, it optimizes both a basis matrix and a parameter matrix. Notably, if the kernel matrix is an identity matrix, the method simplifies to standard NMF. Also provides NMF with Random Effects (NMF-RE) via nmfre(), which estimates a mixed-effects model combining covariate-driven scores with unit-specific random effects together with wild bootstrap inference, and NMF with feed-forward and feedback structure (NMF-FFB) via nmf.ffb(), which fits a two-block input-output model for blind source separation and path analysis. References: Satoh (2025) <doi:10.48550/arXiv.2403.05359>; Satoh (2026) <doi:10.1007/s42081-026-00349-x>; Satoh (2025) <doi:10.48550/arXiv.2512.18250>; Satoh (2026) <doi:10.48550/arXiv.2603.01468>; Satoh and Tokuda (2026) <doi:10.48550/arXiv.2607.27474>; Satoh (2026) <doi:10.1007/s42081-025-00314-0>.

Authors:Kenichi Satoh [aut, cre]

nmfkc_1.0.2.tar.gz
nmfkc_1.0.2.zip(r-4.7-any)nmfkc_1.0.2.zip(r-4.6-any)nmfkc_1.0.2.zip(r-4.5-any)
nmfkc_1.0.2.tgz(r-4.6-any)nmfkc_1.0.2.tgz(r-4.5-any)
nmfkc_1.0.2.tar.gz(r-4.7-any)nmfkc_1.0.2.tar.gz(r-4.6-any)
nmfkc_1.0.2.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
nmfkc/json (API)

# Install 'nmfkc' in R:

install.packages('nmfkc', repos = c('https://ksatohds.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/ksatohds/nmfkc/issues

Pkgdown/docs site:https://ksatohds.github.io

On CRAN:

6.85 score 7 stars 73 scripts 383 downloads 72 exports 0 dependencies

Last updated from:4301c52914. Checks:9 OK. Indexed: yes.

Target	Result	Time
linux-devel	OK	167
source / vignettes	OK	316
linux-release	OK	181
macos-release	OK	186
macos-oldrel	OK	206
windows-devel	OK	140
windows-release	OK	142
windows-oldrel	OK	99
wasm-release	OK	118

Exports:nmf.cluster.criteria nmf.cluster.flow nmf.ffb nmf.ffb.cv nmf.ffb.diagnostics nmf.ffb.DOT nmf.ffb.ecv nmf.ffb.inference nmf.ffb.split nmf.ffb.test nmf.gmm nmf.gmm.inference nmf.gmm.select nmf.gmm.twostage nmf.rrr nmf.rrr.cv nmf.rrr.DOT nmf.rrr.ecv nmf.rrr.heatmap nmf.rrr.inference nmf.rrr.kernel.beta.cv nmf.rrr.rank nmf.rrr.rename nmf.rrr.signed nmf.rrr.signed.ecv nmf.rrr.signed.heatmap nmf.rrr.signed.inference nmf.rrr.signed.rank nmf.rrr.signed.rename nmfkc nmfkc.ar nmfkc.ar.degree.cv nmfkc.ar.DOT nmfkc.ar.latent nmfkc.ar.latent.inference nmfkc.ar.predict nmfkc.ar.stationarity nmfkc.ard nmfkc.bicv nmfkc.class nmfkc.consensus nmfkc.criterion nmfkc.cv nmfkc.denormalize nmfkc.DOT nmfkc.ecv nmfkc.inference nmfkc.kernel nmfkc.kernel.beta.cv nmfkc.kernel.beta.nearest.med nmfkc.kernel.gaussian nmfkc.kernel.gram nmfkc.net nmfkc.net.DOT nmfkc.net.ecv nmfkc.net.inference nmfkc.net.rank nmfkc.normalize nmfkc.rank nmfkc.residual.plot nmfkc.rff.beta.cv nmfkc.rff.positive nmfkc.rff.positive.gram nmfkc.signed nmfkc.signed.cv nmfkc.signed.ecv nmfkc.signed.rank nmfkc.signed.rff nmfkc.signed.rff.gram nmfre nmfre.ecv nmfre.inference

Dependencies:

Introduction to nmfkc

Rendered fromintroduction-to-nmfkc.Rmdusingknitr::rmarkdown

Last update: 2025-11-24
Started: 2025-10-13

Help page	Topics
Extract coefficients from NMF models	coef.nmf coef.nmf.ffb
Extract the covariate-coefficient matrix from an NMF-GMM fit	coef.nmf.gmm
Extract fitted values from NMF models	fitted.nmf fitted.nmf.ffb fitted.nmfae fitted.nmfre
Fitted (responsibility-averaged) reconstruction of an NMF-GMM fit	fitted.nmf.gmm
Sample-clustering quality across ranks	nmf.cluster.criteria
Cluster-flow (alluvial) diagram across a sequence of fits	nmf.cluster.flow
NMF-FFB: non-negative matrix factorization with latent feedback	nmf.ffb
Cross-Validation for NMF-FFB	nmf.ffb.cv
Structural Diagnostics of a Fitted NMF-FFB Feedback Matrix	nmf.ffb.diagnostics
Generate a Graphviz DOT Diagram for an NMF-FFB Model	nmf.ffb.DOT
Choose the number of factors for NMF-FFB by element-wise cross-validation	nmf.ffb.ecv
Bootstrap inference for NMF-FFB (X held fixed)	nmf.ffb.inference
Heuristic Variable Splitting for NMF-FFB	nmf.ffb.split
Test the feed-forward null of an NMF-FFB fit	nmf.ffb.test
Fit NMF-GMM: a Gaussian-mixture latent-class extension of NMF with covariates	nmf.gmm
Statistical inference for an NMF-GMM fit (given basis)	nmf.gmm.inference
Choose the number of mixture components K for NMF-GMM	nmf.gmm.select
Two-stage (adjust-then-cluster) baseline for NMF-GMM	nmf.gmm.twostage
Three-Layer Non-negative Matrix Factorization (NMF-AE)	nmf.rrr
Sample-wise k-fold Cross-Validation for nmfae	nmf.rrr.cv
DOT graph visualization for nmfae objects	nmf.rrr.DOT
Element-wise Cross-Validation for nmfae (Wold's CV)	nmf.rrr.ecv
Heatmap visualization of nmfae factor matrices	nmf.rrr.heatmap
Statistical Inference for NMF-AE Parameter Matrix	nmf.rrr.inference
Optimize kernel beta for nmfae by cross-validation	nmf.rrr.kernel.beta.cv
Rank selection for nmfae (paired rank, concise diagnostics)	nmf.rrr.rank
Rename decoder and encoder bases	nmf.rrr.rename
Signed-Bottleneck NMF-AE: Three-Layer NMF-AE with Signed Bottleneck	nmf.rrr.signed
Element-wise Cross-Validation for Signed-Bottleneck NMF-AE	nmf.rrr.signed.ecv
Heatmap visualization of nmfae.signed factor matrices	nmf.rrr.signed.heatmap
Statistical Inference for Signed-Bottleneck NMF-AE Signed Bottleneck	nmf.rrr.signed.inference
Rank selection for nmfae.signed (paired rank, concise diagnostics)	nmf.rrr.signed.rank
Rename Resp/Cov labels on nmfae.signed objects	nmf.rrr.signed.rename
Optimize NMF with kernel covariates (Full Support for Missing Values)	nmfkc
Construct observation and covariate matrices for a vector autoregressive model	nmfkc.ar
Optimize lag order for the autoregressive model	nmfkc.ar.degree.cv
Generate a Graphviz DOT Diagram for NMF-AR / NMF-VAR Models	nmfkc.ar.DOT
Latent transition matrices of an NMF-VAR model	nmfkc.ar.latent
Bootstrap inference for the latent VAR of an NMF-VAR model	nmfkc.ar.latent.inference
Forecast future values for NMF-VAR model	nmfkc.ar.predict
Stationarity of an NMF-VAR model	nmfkc.ar.stationarity
Automatic relevance determination for NMF rank (experimental)	nmfkc.ard
Bi-cross-validation for NMF rank selection	nmfkc.bicv
Create a class (one-hot) matrix from a categorical vector	nmfkc.class
Consensus-clustering rank selection for NMF (Brunet 2004)	nmfkc.consensus
Compute model selection criteria for a fitted nmfkc model	nmfkc.criterion
Perform k-fold cross-validation for NMF with kernel covariates	nmfkc.cv
Print/plot methods for nmfkc cross-validation objects	nmfkc.cv.methods plot.nmfkc.bicv plot.nmfkc.cv plot.nmfkc.ecv print.nmfkc.bicv print.nmfkc.cv print.nmfkc.ecv
Denormalize a matrix from [0,1] back to its original scale	nmfkc.denormalize
Generate Graphviz DOT Scripts for NMF or NMF-with-Covariates Models	nmfkc.DOT
Perform Element-wise Cross-Validation (Wold's CV)	nmfkc.ecv
Statistical inference for the parameter matrix C (Theta)	nmfkc.inference
Create a kernel matrix from covariates	nmfkc.kernel
Optimize beta of the Gaussian kernel function by cross-validation	nmfkc.kernel.beta.cv
Estimate Gaussian/RBF kernel parameter beta from covariates (supports landmarks)	nmfkc.kernel.beta.nearest.med
Create a Gaussian kernel matrix from covariates	nmfkc.kernel.gaussian
Block-wise Gram accumulation of Nystroem kernel covariates for large N	nmfkc.kernel.gram
Symmetric NMF for networks (tri / bi / signed)	nmfkc.net
Generate a Graphviz DOT Diagram for a Symmetric NMF Network	nmfkc.net.DOT
Element-wise cross-validation for nmfkc.net (upper-triangle folds)	nmfkc.net.ecv
Statistical Inference for Symmetric NMF Parameters	nmfkc.net.inference
Rank selection for nmfkc.net (concise diagnostics)	nmfkc.net.rank
Normalize a matrix to the range [0,1]	nmfkc.normalize
Rank selection diagnostics with graphical output	nmfkc.rank
Plot Diagnostics: Original, Fitted, and Residual Matrices as Heatmaps	nmfkc.residual.plot
Select the bandwidth of random-feature covariates by cross-validation	nmfkc.rff.beta.cv
Positive (non-negative) random features for the Gaussian kernel	nmfkc.rff.positive
Block-wise Gram accumulation of positive random features for large N	nmfkc.rff.positive.gram
NMF-KC with signed covariate matrix	nmfkc.signed
Column-wise k-fold cross-validation for nmfkc.signed	nmfkc.signed.cv
Element-wise cross-validation for nmfkc.signed	nmfkc.signed.ecv
Rank selection for nmfkc.signed (concise diagnostics)	nmfkc.signed.rank
Random Fourier Features for nmfkc.signed()	nmfkc.signed.rff
Block-wise Gram accumulation of Random Fourier Features for large N	nmfkc.signed.rff.gram
Non-negative Matrix Factorization with Random Effects	nmfre
NMF-RE element-wise cross-validation for rank selection	nmfre.ecv
Statistical inference for the coefficient matrix C from NMF-RE	nmfre.inference
Plot clustering-quality criteria across a sequence of fits	plot.nmf.cluster.criteria
Plot a cluster-flow (alluvial) diagram	plot.nmf.cluster.flow
Plot method for nmf.gmm objects	plot.nmf.gmm
Plot method for nmf.gmm.select objects	plot.nmf.gmm.select
Plot a rank-selection (nmf.rank) object	plot.nmf.rank
'plot.nmfae' displays the convergence trajectory of the objective function across iterations. The title shows the achieved R^2.	plot.nmfae
Plot method for objects of class 'nmfkc'	plot.nmfkc
Plot the latent dynamics of an NMF-VAR model	plot.nmfkc.ar.latent
Plot the stationarity bracket of an NMF-VAR model	plot.nmfkc.ar.stationarity
Plot method for nmfkc.ard objects	plot.nmfkc.ard
Plot a consensus rank-selection (nmfkc.consensus) object	plot.nmfkc.consensus
Plot method for nmfkc.DOT objects	plot.nmfkc.DOT
Plot method for nmfkc.signed (convergence)	plot.nmfkc.signed
Plot convergence diagnostics for NMF models	plot.nmf.ffb plot.nmfre
Plot method for nmfre.ecv objects	plot.nmfre.ecv
Class assignments / responsibilities from an NMF-GMM fit	predict.nmf.gmm
Prediction method for objects of class 'nmfkc'	predict.nmfkc
Predict method for nmfkc.signed	predict.nmfkc.signed
Predict method for nmfre objects	predict.nmfre
Print method for fitted NMF models	print.nmf
Print method for nmf.cluster.criteria objects	print.nmf.cluster.criteria
Print method for nmf.cluster.flow objects	print.nmf.cluster.flow
Print a calibrated test of the feed-forward null	print.nmf.ffb.test
Print an NMF-GMM fit	print.nmf.gmm
Print method for nmf.gmm.select objects	print.nmf.gmm.select
Print method for NMF inference objects	print.nmf.inference
Print method for rank-selection (nmf.rank) objects	print.nmf.rank
Print method for nmf.rrr element-wise CV results	print.nmfae.ecv
Print method for nmfkc.ar.latent objects	print.nmfkc.ar.latent
Print method for nmfkc.ar.latent.inference objects	print.nmfkc.ar.latent.inference
Print method for nmfkc.ar.stationarity objects	print.nmfkc.ar.stationarity
Print method for nmfkc.ard objects	print.nmfkc.ard
Print method for nmfkc.consensus objects	print.nmfkc.consensus
Print method for nmfkc.DOT objects	print.nmfkc.DOT
Print method for nmfkc.gram	print.nmfkc.gram
Print method for nmfre.ecv objects	print.nmfre.ecv
Print method for summary.nmf.ffb objects	print.summary.nmf.ffb
Print method for summary.nmf.gmm objects	print.summary.nmf.gmm
Print method for 'summary.nmfkc' objects	print.summary.nmfkc
Print method for summary.nmfkc.inference objects	print.summary.nmfkc.inference
Print method for summary.nmfkc.net objects	print.summary.nmfkc.net
Print method for summary.nmfkc.net.inference objects	print.summary.nmfkc.net.inference
Print method for summary.nmfkc.net.signed objects	print.summary.nmfkc.net.signed
Print method for summary.nmfkc.signed	print.summary.nmfkc.signed
Extract residuals from NMF models	residuals.nmf residuals.nmf.ffb residuals.nmfae residuals.nmfre
Residuals from an NMF-GMM fit	residuals.nmf.gmm
Summary method for nmf.ffb objects	summary.nmf.ffb
Summary of an NMF-GMM fit	summary.nmf.gmm
Summary method for objects of class 'nmfkc'	summary.nmfkc
Summary method for nmfkc.inference objects	summary.nmfkc.inference
Summary method for nmfkc.net objects	summary.nmfkc.net
Summary method for nmfkc.net.inference objects	summary.nmfkc.net.inference
Summary method for nmfkc.net.signed objects	summary.nmfkc.net.signed
Summary method for nmfkc.signed	summary.nmfkc.signed
Summary method for objects of class 'nmfre'	print.summary.nmfre summary.nmfre

Package: nmfkc 1.0.2

nmfkc: Non-Negative Matrix Factorization with Kernel Covariates

Citation

Development and contributors

Readme and manuals

Help Manual

Usage by other packages (reverse dependencies)