Skip to content

exametrika v1.14.0

Choose a tag to compare

View all tags
@kosugitti kosugitti released this 14 Jun 23:46
· 92 commits to main since this release
v1.14.0
02a5ae2

Summary

This is a minor release that responds to an R Journal review request and
fixes several bugs. There are no breaking changes; all deprecated
names from previous releases continue to work unchanged.

The headline change is that plot.exametrika() now actually forwards the
graphical parameters you pass through ... (e.g. las, pch, cex,
col) to every plot type — previously they were silently ignored even
though the help page documented them.

Improvements

  • plot.exametrika() now forwards graphical parameters supplied via
    ... to every plot type (R Journal review request).     The ...
    argument documented on the plot.exametrika() help page was previously
    captured but never passed on, so standard graphical parameters were
    silently ignored. For example,

    plot(result.LRA, type = "IRP", items = 1:4, nc = 2, nr = 2,
     las = 2, pch = 16)

    changed neither the axis-label orientation (las) nor the plotting
    symbol (pch). User-supplied parameters are now forwarded consistently
    to the underlying base R plotting calls (plot, barplot, image,
    lines, curve) for all plot types, including manually drawn
    axes    , so las, cex.axis, and similar parameters take effect on
    every axis. User values take precedence over the package defaults, so
    xlab, ylab, main, etc. can be overridden. No new dependency is
    added; the package remains base-graphics only.

Bug fixes

  • Binary IRM: crash on real data with missing values.
    Biclustering_IRM() on a binary exametrika data object containing
    missing cells crashed inside rmultinom() ("NA in probability
    vector") because the -1 missingness marker from dataFormat() was
    aggregated directly. The masked matrix is now used downstream.
    Reproduced on HCI (651×20, 1.5% missing) and SAT12 (600×32 binarised,
    0.4% missing).

  • Graphical Lasso: graceful handling of numerical divergence. When
    the inner block coordinate descent of Glasso() produced non-finite
    values (typically near-singular polychoric matrices in the N <= p
    regime with very small lambdas), the previous implementation crashed
    with "missing value where TRUE/FALSE needed" and discarded all work
    along the lambda grid. Glasso() now warns, breaks the lambda search,
    and returns the best solution found so far; skipped lambdas are marked
    NA in the result path.

  • Cosmetic fixes. IRT Test Information Function plot title typo
    ("Informaiton" → "Information"), the Biclustering array-plot panel
    title ("Clusterd Data" → "Clustered Data"), and four user-facing
    message typos. The ggExametrika::plotArray_gg() argument names that
    inherited the Clusterd typo are corrected in the coordinated
    ggExametrika 1.1.1 release.

Install

install.packages("exametrika")
# or the development version:
# remotes::install_github("kosugitti/exametrika")

Full changelog: https://kosugitti.github.io/exametrika/news/

Assets 2
Loading