mc.meas.iter: Summary of a predictor in mc.agg object

In wilsontom/FIEmspro: Flow In-jection Electrospray Mass Spectrometry Processing: \\ data processing, classification modelling and variable selection in metabolite fingerprinting

Description

Convenience function to output statistics related to accuracy, AUC or margins at each iteration for one model or a selection of models contained in a mc.agg object (see details mc.agg).

Usage

mc.meas.iter(mc.obj, lmod = NULL,type="acc",nam="Model")

Arguments

mc.obj	mc.agg object - See details mc.agg
lmod	List of models to be considered - Default: all models
type	Predictor type - Can be either acc (accuracy), auc (AUC), mar (margin or equivalent)
nam	List of names to be used in the result - Names given here corresponds to the column name of mc.obj$cldef

Value

Data frame containing statistic of interest at each iteration.

Author(s)

David Enot [e-mail]

See Also

mc.agg

Examples

data(iris)
dat=as.matrix(iris[,1:4])
cl=as.factor(iris[,5])
lrnd=sample(1:150)[1:50]
cl[lrnd]=sample(cl[lrnd])  ## add a bit of misclassification for fun
pars   <- valipars(sampling = "cv",niter = 10, nreps=4)
dat1=dat.sel1(dat,cl,pwise="virginica",mclass=NULL,pars=pars)

res1=lapply(dat1,function(x) accest(x,clmeth="lda"))
res2=lapply(dat1,function(x) accest(x,clmeth="randomForest",ntree=50))

## Aggregate res1 and res2
mc=mc.agg(res1,res2)

## AUC in each model
auc.iter<-mc.meas.iter(mc,type="auc",nam=c("DisId","Alg"))
## Plot them
boxplot(auc.iter)
## Print on the screen
print(auc.iter)

wilsontom/FIEmspro documentation built on May 4, 2019, 6:28 a.m.