Plots: Advanced Plots

In matloff/qeML: Quick and Easy Machine Learning Tools

Advanced Plots

Description

Miscellaneous specialized plots.

Usage

plotPairedResids(data,qeOut) 
plotClassesUMAP(data,classVar) 
qeFreqParcoord(dataName,k=25,opts=NULL)
qePlotCurves(curveData,xCol=1,yCol=2,grpCol=3,
   xlab=names(curveData)[xCol],ylab=names(curveData)[yCol],
   loess=TRUE,legendTitle,legendSpace=1.1,legendPos='topright')

qePlotCurves(curveData,xCol=1,yCol=2,grpCol=3,
   xlab=names(curveData)[xCol],ylab=names(curveData)[yCol],loess=TRUE,
   legendTitle=names(curveData)[grpCol],legendSpace=1.1,legendPos="topright",
   wide=FALSE,wideTimeColName=NULL,wideTimeColPresent=NULL,
      wideTimeColBase=1:nrow(curveData),wideGrpColName=NULL,
      wideValueColName=NULL)
qeMittalGraph(dataMitt,xlab="x",ylab="y",legendTitle="curve",loess=TRUE) 

Arguments

data	A data frame or equivalent..
dataMitt	A data frame or equivalent. "X" and "Y" columns, followed by a group column, an R factor.
qeOut	An object returned from one of the qe-series predictive functions..
classVar	Name of the column containing class information.
dataName	Quoted name of a data frame.
k	Number of nearest neighbors.
opts	Options to be passed to discparcoord.
curveData	Data to be plotted.
xCol	Column name or number containing "X".
yCol	Column name or number containing "Y".
grpCol	Column name or number containing group name, a character vector or factor.
xlab	X-axis label.
ylab	Y-axis label.
loess	If TRUE, do loess smoothing within each group.
legendTitle	Legend title.
legendSpace	Factor by which to expand vertical space, to accommodate a top-situated legend.
legendPos	Position of legend within plot.
curveData	A data frame, "X" values in column 1.
xlab	Label for X-axis.
ylab	Label for Y-axis.
wide	TRUE if curveData is in wide format.
wideTimeColName	Name to be used for "X"-axis.
wideTimeColPresent	If TRUE, a time column already exists.
wideTimeColBase	"Time" values for each group.
wideGrpColName	Group name.
wideGrpValueName	"Y" value name.

Details

The plotPairedResids function plots model residuals against pairs of features, for example for model validation. Pairs are chosen randomly.

The function qeFreqParcoord is a qeML interface to the cdparcoord package.

The function qePlotCurves plots X-Y curves for one or more groups. Within each group, the (X,Y) pairs are plotted, possibly with loess smoothing. Input data format long by default, but can be wide.

The function qeMittalGraph is similar to qePlotCurves, except that it displays multiplicative change over the X-axis. All curves start at height 1.0. (There may be some exceptions to this if loess is TRUE.) The X-axis could be time or some model parameter, say in graphing prediction accuracy against number of nearest neighbors for different datasets.

Author(s)

Norm Matloff

Examples

## Not run: 

data(pef)
linout <- qeLin(pef,'wageinc')
plotPairedResids(pef,linout)

data(lsa)
# plot LSAT score against undergradute GPA, for each law school cluster
# (reputed quality of the law school)
qePlotCurves(lsa,6,5,9,legendSpace=1.35)

data(currency)
curr <- cbind(1:nrow(currency),currency)
names(curr)[1] <- 'weeknum'
qePlotCurves(currency,wide=TRUE,wideTimeColName='weeknum',
   wideTimeColPresent=FALSE,wideGrpColName='country',wideValueColName='rate')
qeMittalGraph(curr,'weeknum','rate','country')
# Canadian dollar and pound in one cluster, and franc, mark and
# yen in another



## End(Not run)

matloff/qeML documentation built on Dec. 15, 2024, 10:15 a.m.