allows for more control over graphical parameters.
accounts for NULL stringsAsFactors option.
appropriately drops levels when they are based on a factor.
dark option to more easily set text colors for dark backgrounds.
standardizes color codes before passing them to par.
avoids overwriting par color settings.
better handles arguments outside of options when options is specified.
treats logical y variables as binary rather than factors.
check equivalency of expression outputs within splot.bench.
handles input with as.matrix but not as.data.frame methods.
adds a cutoff option and relative timings to splot.bench for better display, and includes more in its return.
splot.bench now evaluates within the global environment so loaded packages can be used.
specify how colors are interpolated with color.summary.
calculate average colors with the splot.colormean function.
fill the area under each density lines with its color with density.fill = TRUE (default), adjust the fill’s opacity with density.opacity, and further control the densities with arguments in density.args (which replaces the old bw and adj arguments).
broadens colorby applicability and flexibility.
tweaks default spacing rules to maximize plot areas.
aligns axis label specification for density with other types.
better retains or infers names for multiple variables.
avoids an error when colorby’s by has missing values.
corrects several colorby assignment issues.
handles mixed-type multiple ys by converting non-numeric variables to numeric when other numeric variables are included.
maintains variable ordering when displaying counts of categorical ys.
includes colorby variable name in split note when split.
assign colors in the scale of numeric values with splot.color’s grade argument.
makes a legend for colorby in more cases.
allows renaming and reordering of character or factor y levels.
more broadly retains character and factor level orders.
more broadly accounts for missing values.
better handles colorby x aggregation.
splot.color returns shorter color codes when opacity is 1.
corrects application of adjustments in splot.color when no sampling method is used.
corrects connected line paths when sparse.
calculates frequencies of y levels within by and between splits.
corrects some colorby x and by assignments.
preserves color assignments when colorby has multiple arguments, and some levels of by are empty.
avoids error when a level of split is empty.
allows splot.color sampling to be turned off.
allows splot.bench to handle single runs.
enables more sorting options.
correctly does not display error bars when error is set to FALSE.
corrects the x-axis range of scatter plots when x is logical.
corrects bet reassignment given multiple ys.
place the legend interactively.
adjust color opacity, and assign colors in gradients.
split by specified values.
display predicted probability lines in scatter plots with 2-level ys.
allows for non-numeric y variables.
general improvements in automatic settings.
more reliable connection between data and line colors, types, and widths.
provides more control over the legend and colorby.
improved reading of the y and su arguments.
improved bar plot y-axis labels.
now appropriately recycles short variables.
avoids an error when levels of by are missing within between splits.
sort text x variables by their y values.
split by a specified number of segments.
cleared up sub/sud related mixups.
level ordering now applies to between variables.
improved formula parsing; expressions should be better recognized as arguments.
rearranged arguments to be more like other functions; variables other that y must be named, or entered as part of the y formula.
multiple variables entered as y can now be renamed/ordered with an mv entry in levels.
corrected y-axis labeling for bar plots when autori is true.
custom notes now overwrite all other note elements.
specified x levels are now appropriately applied when mv.as.x is true.
standardized lty behavior across types, and prevented it from causing errors.
fixed an issue with longer vectors entered directly as arguments.
multiple y variables with the same name are now treated as different.
added colorby argument, offering a different way to assign colors.
added utility functions: splot.color to assign colors by group, and splot.bench to plot the run-time of expressions.
allows for text as x-axis labels in scatter plots.
changed the way variables are evaluated to accommodated calls to splot from within functions.
improved handling of between variables with uneven levels.
improved variable/level length handling.
improved handling of prediction line errors and some fatal errors.
bar and line plots no longer skip levels of x with only one level of by.
corrected occasional mismatch between x labels and data.