Package: prinsurf 2.0

prinsurf: Principal Surface Biplots

Fits principal surfaces (Hastie and Stuetzle, 1989) and constructs smooth, predictive biplot axes for them by gradient flow: each variable's calibrated axis is the steepest-ascent trajectory of its fitted surface coordinate, which is smooth by construction, crosses the variable's contour lines orthogonally, and reduces to the linear principal-component-analysis biplot axis when the surface is flat. Variables whose fitted value has an interior extremum (closed contours) are detected as non-monotone and the axis is withheld in favour of contour-based reading. The implementation is base-R only.

Authors:Raeesa Ganey [aut, cre]

prinsurf_2.0.tar.gz
prinsurf_2.0.zip(r-4.7)prinsurf_2.0.zip(r-4.6)prinsurf_2.0.zip(r-4.5)
prinsurf_2.0.tgz(r-4.6-any)prinsurf_2.0.tgz(r-4.5-any)
prinsurf_2.0.tar.gz(r-4.7-any)prinsurf_2.0.tar.gz(r-4.6-any)
prinsurf_2.0.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION
card.svg |card.png
prinsurf/json (API)

# Install 'prinsurf' in R:
install.packages('prinsurf', repos = c('https://raeesaganey91.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/raeesaganey91/prinsurf/issues

On CRAN:

Conda:

4.30 score 2 scripts 180 downloads 5 exports 0 dependencies

Last updated from:190eef98bb. Checks:7 WARNING, 2 OK. Indexed: yes.

TargetResultTimeFilesSyslog
linux-devel-x86_64WARNING176
source / vignettesOK196
linux-release-x86_64WARNING130
macos-release-arm64WARNING105
macos-oldrel-arm64WARNING94
windows-develWARNING66
windows-releaseWARNING63
windows-oldrelWARNING80
wasm-releaseOK89

Exports:axis_predictive_errorkink_maxpredictivityprinsurfpsaxis

Dependencies:

Principal surface biplots with prinsurf
Introduction | A curved sheet | A half-sphere cap: deferral in action | Iris: a real dataset with groups | How well can the display be read? | Two reading rules | Practical notes | References
Rendered fromprinsurf.Rmdusingknitr::rmarkdown

Last update: 2026-07-02
Started: 2026-06-26

Smooth predictive axes for principal surface biplots
Setup | The axis as a trajectory of the fitted response | Smoothness and orthogonality | Reduction to the linear biplot | Monotonicity: two criteria | Reading and prediction | Which fidelity measure, and why $[0,1]$ fails | What is established, and what is not | A robust discrete estimator | Algorithm | Estimation | References
Rendered fromgradient-flow-theory.Rmdusingknitr::rmarkdown

Last update: 2026-07-02
Started: 2026-07-02