[R] argument "x" is missing, with no default - Please help find argument x

[dunner]

R programming question, not machine learning, although that's the content.

Apologies to all for whom the following code is eye-burning. I am using
foreach() to run a simulation on a randomForest model (actually conditional
randomForest ... "party" package). The simulation is in two dimensions.
examining how "mtry" and "ntrees" are related in terms of predictive
accuracy in ten-fold cross validation.
My problem is one of functional programming. The "loops" for simulation are
functionalised so they can be passed to foreach and bundled of to my 4
cores. However, I'm making a mess of passing in arguments - I think. I get
the following error message :

Error in toploop(ntrees = ntrees, ml.frame = ml.frame) :
task 2 failed - "argument "x" is missing, with no default"

as if it couldn't see the arguments.

Long segment of actual code below.

Thank you for your time and patience.

Ross

<code>

bottomloop<-function(i=i, mtry=mtry, rownumber=rownumber, ntrees=ntrees,
shuffled=shuffled) {
require(party)
require(epiR)
test<-rownumber[[i]]
train<-shuffled[shuffled!=test]
cv.train <- ml.frame[train,]
cv.evaluate <- ml.frame[test,]
cv.model <- cforest(as.factor(remitter)~., data=cv.train,
control=cforest_control(mtry=mtry, ntree=ntrees, mincriterion=0.95))
cv.evaluate$prediction <- predict(cv.model, newdata=cv.evaluate)
obj.pred<-predict(cv.model)
pred.table<-table(cv.evaluate$prediction, cv.evaluate$remitter)

truepos<-pred.table[1,1]
trueneg<-pred.table[2,2]
falsepos<-pred.table[2,1]
falseneg<-pred.table[1,2]
sumvar<-epi.tests(c(truepos, falsepos , falseneg, trueneg), verbose=T)
epi.tests(c(truepos, falsepos , falseneg, trueneg), verbose=F)
# Calculate the overall accuracy.
cv.evaluate$correct <- cv.evaluate$prediction == cv.evaluate$remitter
mymean<-mean(cv.evaluate$correct)
retlist<-data.frame(mtry, sumvar$se, sumvar$sp, mean(cv.evaluate$correct))
return(retlist)
}

subloop<- function(mtry=mtry, ml.frame=ml.frame, ntrees=ntrees) {
nfolds<- 10
# shuffle the numbers and divide into 10 groups
numberOfRows<-dim(ml.frame)[1]
lengthOfDiv<-numberOfRows/nfolds
shuffled<-sample(c(1:numberOfRows), numberOfRows, replace=F)
rownumber<-split(shuffled, 1:nfolds)
#combine mymean into my vec
myvec<-foreach (i = 1:length(rownumber), .combine="rbind") %dopar%
bottomloop(i, mtry = mtry, rownumber=rownumber, ntrees=ntrees,
shuffled=shuffled)
#print(myvec)
return(myvec)
}

toploop <- function (ntrees=ntrees, ml.frame=ml.frame) {
require(foreach)
require(ggplot2)
require(reshape)
require(foreign)
mtry = c(5:25)
ddd<-foreach(mtry, .combine="rbind") %dopar% subloop (mtry=mtry,
ml.frame=ml.frame, ntrees=10)
ddd<-as.data.frame(ddd)
mdd<-melt(ddd, id="mtry")
g <- ggplot(mdd, aes(mtry, value))
pdf(file=paste(ntrees, "-trees.pdf"))
g + geom_point(aes(colour=as.factor(variable))) +
geom_smooth(aes(group=as.factor(mdd$variable),
colour=as.factor(mdd$variable)), size=1.2)
dev.off()
write.csv(file=paste(ntrees, "-trees.csv"))
}

ntrees=c(1,2,3)
foreach(ntrees, .combine="cbind" ) %dopar% toploop (ntrees=ntrees,
ml.frame=ml.frame)



> foreach(ntrees, .combine="cbind" ) %dopar% toploop (ntrees=ntrees,
> ml.frame=ml.frame)

bottomloop<-function(i=i, mtry=mtry, rownumber=rownumber, ntrees=ntrees,
shuffled=shuffled) {
require(party)
require(epiR)
test<-rownumber[[i]]
train<-shuffled[shuffled!=test]
cv.train <- ml.frame[train,]
cv.evaluate <- ml.frame[test,]
cv.model <- cforest(as.factor(remitter)~., data=cv.train,
control=cforest_control(mtry=mtry, ntree=ntrees, mincriterion=0.95))
cv.evaluate$prediction <- predict(cv.model, newdata=cv.evaluate)
obj.pred<-predict(cv.model)
pred.table<-table(cv.evaluate$prediction, cv.evaluate$remitter)

truepos<-pred.table[1,1]
trueneg<-pred.table[2,2]
falsepos<-pred.table[2,1]
falseneg<-pred.table[1,2]
sumvar<-epi.tests(c(truepos, falsepos , falseneg, trueneg), verbose=T)
epi.tests(c(truepos, falsepos , falseneg, trueneg), verbose=F)
# Calculate the overall accuracy.
cv.evaluate$correct <- cv.evaluate$prediction == cv.evaluate$remitter
mymean<-mean(cv.evaluate$correct)
retlist<-data.frame(mtry, sumvar$se, sumvar$sp, mean(cv.evaluate$correct))
return(retlist)
}

subloop<- function(mtry=mtry, ml.frame=ml.frame, ntrees=ntrees) {
nfolds<- 10
# shuffle the numbers and divide into 10 groups
numberOfRows<-dim(ml.frame)[1]
lengthOfDiv<-numberOfRows/nfolds
shuffled<-sample(c(1:numberOfRows), numberOfRows, replace=F)
rownumber<-split(shuffled, 1:nfolds)
#combine mymean into my vec
myvec<-foreach (i = 1:length(rownumber), .combine="rbind") %dopar%
bottomloop(i, mtry = mtry, rownumber=rownumber, ntrees=ntrees,
shuffled=shuffled)
#print(myvec)
return(myvec)
}

toploop <- function (ntrees=ntrees, ml.frame=ml.frame) {
require(foreach)
require(ggplot2)
require(reshape)
require(foreign)
mtry = c(5:25)
ddd<-foreach(mtry, .combine="rbind") %dopar% subloop (mtry=mtry,
ml.frame=ml.frame, ntrees=100)
ddd<-as.data.frame(ddd)
mdd<-melt(ddd, id="mtry")
g <- ggplot(mdd, aes(mtry, value))
pdf(file=paste(ntrees, "-trees.pdf"))
g + geom_point(aes(colour=as.factor(variable))) +
geom_smooth(aes(group=as.factor(mdd$variable),
colour=as.factor(mdd$variable)), size=1.2)
dev.off()
write.csv(file=paste(ntrees, "-trees.csv"))
}

ntrees=c(1,2,3)
foreach(ntrees, .combine="cbind" ) %dopar% toploop (ntrees=ntrees,
ml.frame=ml.frame)

### following error ###

> foreach(ntrees, .combine="cbind" ) %dopar% toploop (ntrees=ntrees,
> ml.frame=ml.frame)
Error in toploop(ntrees = ntrees, ml.frame = ml.frame) :
task 2 failed - "argument "x" is missing, with no default"

</code>


--
View this message in context: http://r.789695.n4.nabble.com/argument-x-is-missing-with-no-default-Please-help-find-argument-x-tp4633551.html
Sent from the R help mailing list archive at Nabble.com.

______________________________________________
R-h...@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.