mc.summary: Summary of multiple classifiers objects

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

Description

Convenience function to output statistics related to accuracy, AUC and margins for a selection of models. If sortDis=TRUE, results are grouped by discrimination task (value contained in DisId column of mc.obj$cldef). If sortDis=FALSE, results are grouped by classifier algorithm (value contained in column AlgId of mc.obj$cldef).

Usage

mc.summary(mc.obj,lmod=NULL,sortDis=TRUE)

## Default S3 method:
mc.summary(mc.obj, lmod = NULL, sortDis = TRUE)

Arguments

mc.obj	mc.agg object
lmod	List of models to be considered - Default: all of them
sortDis	Should the results be sorted by discrimination task? If FALSE, results are group by classifier techniques

Value

mc.summary object:

res	List of results
cltask	Discrimination task(s) or classification algorithm(s) used.
title	Title for printing function.

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])
pars   <- valipars(sampling = "cv",niter = 2, nreps=10)
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)

## Sort results by discrimination task
mc.summary(mc)

## Sort results by algorithm
mc.summary(mc,sortDis=FALSE)

## See what is in
names(mc.summary(mc))

## Print results in a file
## Not run: print(mc.summary(mc,sortDis=FALSE),digits=2,file="tmp.csv")

tonedivad/FIEmspro documentation built on May 31, 2019, 6:20 p.m.