Releases: pbreheny/visreg
Releases · pbreheny/visreg
Version 2.7.0
- Changed: Plots are now constructed in a consistent order across all varieties (lattice, base, gg, overlay, etc.), with lines on top (fixes #50)
- New: Option to change the default layering with top="points"
- Fixed: Resolved namespace issue with factor plots using lattice
- Fixed: Explicitly convert to factor (no longer automatic as of R 4.0)
- Fixed: strip.names and rug colors now work correctly for ggplot (fixes #85)
Version 2.6.1
Version 2.6.0
- New:
visreg()
,visreg2d()
now acceptdata=
option for explicit data passing - Fixed: coxph models now work when
update()
is required (#76) - Fixed: Improved formula parsing (now uses
all.vars
) - Changed:
plot.visreg2d()
now returns transformation matrix for persp plots - Internal: TravisCI now used for automatic testing
- Internal: More consistent handling of matrix outcome (mlm)
Version 2.5-1
Due to changes in other packages, visreg needed some updates to maintain compatibility:
- Fixed: Restoring compatibility with current version of quantreg
- Fixed: Restoring compatibility with current version of survey
Version 2.5-0
- New: Overlay now works for gg plots
- New: gg option for visreg2d
- New: Improved support for betareg models
- New: Support for polr models
- Changed: visreg2d now split into calculation and plotting functions
- Changed: vignette now HTML instead of PDF
- Fixed: Formula parser now handles nested parentheses
- Fixed: Multi-reponse models with missing residuals() methods
- Fixed: Rugs now work with gg plots
- Fixed: Formula parsing for strata() in coxph models
Version 2.4-1
- New: Can now specify reference level for contrast plots by including
x
variable incond
list (pull request #33, thank you @jealie!) - Changed: Default for transformed outcomes now rug=2
- Changed: Default color for persp plot
- Changed: visreg(fit) no longer plots constant objects
- Changed: Appearance of strip names for numeric 'by' variables
- Fixed: plot.visreg wasn't plotting gg objects by default
- Fixed: Bug in formula parsing w/ complicated interactions
- Fixed: Bugs in ggFactorPlot
- Fixed: print.cond warning using warning() instead of printing text
Version 2.4-0
- New: gg=TRUE works for regular plots, not just cross-sectional plots
- New: gg=TRUE returns a gg object, which can then be manipulated further
- Fixed: bug in coxph models with type='contrast'