Error in train.default(x, y, weights = w, ...) : Stopping

[Lan131]

Hello, This might be the wrong place for this question. My errors is

Error in train.default(x, y, weights = w, ...) : Stopping In addition: There were 50 or more warnings (use warnings() to see the first 50)

All I did was modify your regression example and combine it with your tuning example, I have bolded the code I changed:

`train=read.csv("path") trainIndex <- createDataPartition(train[,1], p = .7, list = FALSE, times = 1) DataTrain <- train[ trainIndex,] DataTest <- train[-trainIndex,]

tc <- trainControl(method = "boot", number = 5, allowParallel = T, verboseIter = T)

parameters <- data.frame(parameter = c("layers", "bp.learnRate", "darch.unitFunction"), class = c("character", "numeric", "character"), label = c("Network structure", "Learning rate", "unitFunction"))

grid <- function(x, y, len = NULL, search = "grid") { df <- expand.grid(layers = c("c(0,20,0)","c(0,10,10,0)","c(0,10,5,5,0)"), bp.learnRate = c(1,2,5,10))

df[["darch.unitFunction"]] <- rep(c("c(tanhUnit, linearUnit)", "c(tanhUnit, tanhUnit, linearUnit)", "c(tanhUnit, tanhUnit, tanhUnit, linearUnit)"), 4)

df }

darch <- train(Target ~ ., data = DataTrain, tuneLength = 12, trControl = tc, preProc.targets = T, method = darchModelInfo(parameters, grid), preProc = c("center", "scale"), darch.isClass=F, darch.numEpochs = 15, darch.batchSize = 30, xvalid=DataTest[,3: (ncol(DataTest)-1)],yvalid=DataTest[,2])

data is 200 rows

``

[Lan131]

I can confirm that I am using the correct version of darch.

[saviola777]

I assume you closed this accidentally, or is your problem fixed? I will look into this when I find time, pretty busy atm, sorry.

[saviola777]

I'm not sure if the caret train function supports regression optimization at the moment. The main problem here was that you passed darch.isClass which is not allowed since it was set internally according to the caret parameter classProbs. I have disabled this for now, but I will need to run more tests to see if the regression generally works with caret.