train function does not work the same in current version

[Arcticgrayling]

Coursera Class is running code from class website:
https://github.com/DataScienceSpecialization/courses/blob/master/08_PracticalMachineLearning/016preProcessingPCA/index.Rmd

train function gives error with caret version 6.0-71. problem goes away with older versions.

modelFit <- train(training$type ~ .,method="glm",data=trainPC)

Error in [.data.frame(data, , all.vars(Terms), drop = FALSE) :

undefined columns selected

In addition: There were 26 warnings (use warnings() to see them)

Warnings() gives:

warnings()

Warning messages:

1: glm.fit: fitted probabilities numerically 0 or 1 occurred

2: glm.fit: fitted probabilities numerically 0 or 1 occurred

If I revert to older version of caret 6.0-58 error goes away.
Just Warnings with version 6.0-70.

CODE:
llibrary(caret); library(kernlab); data(spam)
inTrain <- createDataPartition(y=spam$type,
p=0.75, list=FALSE)
training <- spam[inTrain,]
testing <- spam[-inTrain,]

preProc <- preProcess(log10(training[,-58]+1),method="pca",pcaComp=2)
trainPC <- predict(preProc,log10(training[,-58]+1))
modelFit <- train(training$type ~ .,method="glm",data=trainPC)

Preprocessing with PCA

testPC <- predict(preProc,log10(testing[,-58]+1))
confusionMatrix(testing$type,predict(modelFit,testPC))

Alternative (sets # of PCs)

         modelFit <- train(training$type ~ .,method="glm",preProcess="pca",data=training)
         confusionMatrix(testing$type,predict(modelFit,testing))

[topepo]

You shouldn't use the data set name on the LHS of the formula. The formula interface should be used when the variables are in columns of the object that the data argument refers to.

If type is not in training and there are only numeric variables in trainPC, then you should use the non-formula method:

modelFit <- train(x = trainPC, y = training$type,method="glm")

Does that work?

[Arcticgrayling]

That does work, thanks. I'll let the Coursera folks know.

[ireliatt]

Hi, what about this line:

modelFit <- train(training$type ~ .,method="glm",preProcess="pca",data=training)

I deleted data=training but it still doesn't work...