Creates Condvis Shiny app
condvis(
data,
model = NULL,
response = NULL,
sectionvars = NULL,
conditionvars = NULL,
predsInit = NULL,
pointColor = c("steelblue", "grey0"),
cPlotPCP = FALSE,
cPlotn = 1000,
orderConditionVars = "default",
threshold = 1,
thresholdmax = NULL,
linecols = NULL,
showsim = NULL,
theta3d = 45,
phi3d = 20,
dataplot = "pcp",
tours = NULL,
predictArgs = NULL,
xlim = NULL,
ylim = NULL,
zlim = NULL,
density = FALSE,
showdata = density == FALSE,
displayHeight = 950
)
the dataset used for the fit. Should not have NAs for response, sectionvars or conditionvars.
A fitted model or list of models. May be NULL.
name of response variable. If null, tries to extract from model.
names of sectionvars. If null, extracts from data.
names of condition vars. If null, extracts from data.
Optionally provide starting value for some predictors. Defaults to medoid.
a color, or the name of variable to be used for coloring. If the named variable is numeric, it is first converted to a factor with 3 levels.
if TRUE, conditionplots are drawn as a single PCP (for more than two conditionvars)
Defaults to 1000. Shows a sample of this number of points in conditionplots. Non-numeric values are ignored.
If supplied, a function to order the Condition Vars
used for similarity weights, defaults to 1.
maximum value allowed of threshold.
vector of colors to be used for fits
if TRUE, shows sim in conditionplots with points/lines. Defaults to TRUE with 150 or fewer cases.
Angles defining the viewing direction for 3d surface. theta3d
gives the azimuthal direction and phi3d
the colatitude. See
persp
.
"pcp" or "pairs". Used when there is no response, or more than two sectionvars.
A list of pre-calculated tours
a list with one entry per fit, giving arguments for CVpredict
passed on to sectionplot
passed on to sectionplot
passed on to sectionplot
default FALSE. Use TRUE if model is a density function.
defaults to density==TRUE
. If FALSE, data on section not shown.
supply a value for the display height
fit <- lm(mpg ~ wt+hp+am, data=mtcars)
if(interactive()){
condvis(mtcars,fit, response="mpg",sectionvars="wt", conditionvars=c("am", "hp"), pointColor ="red")
}