Train Random Forest with Caret Package (R)
4. Bootstrapping (Repetitive Sampling)
1. No Cross Validation / Bootstrapping
2. Cross Validation with Manual Fine Tuning
mtry <- tuneRF(dev[, -1], dev[,1], ntreeTry=500, stepFactor=1.5,improve=0.01, trace=TRUE, plot=TRUE)
best.m <- mtry[mtry[, 2] == min(mtry[, 2]), 1]
set.seed(825)
trained1 <- train(dev[, -1], dev[,1], method = "rf", ntree=100, tuneGrid=data.frame(mtry=best.m), trControl = trainControl(method = "none"), importance = TRUE)
2. Cross Validation with Manual Fine Tuning
sqtmtry<- round(sqrt(ncol(mydata) - 1))
rfGrid <- expand.grid(mtry = c(round(sqtmtry / 2), sqtmtry, 2 * sqtmtry))
ctrl <- trainControl(method = "cv", classProbs = TRUE, summaryFunction = twoClassSummary, number = 3)
set.seed(2)
trained2<- train(Y ~ . , data = mydata, method = "rf", ntree = 500, tuneGrid = rfGrid, metric = "ROC",
trControl = ctrl, importance = TRUE)
3. Cross Validation with Automatic Fine Tuning
set.seed(2)
trained3 <- train(Y ~ . , data = mydata, method = "rf", ntree = 500, tunelength = 10, metric = "ROC", trControl = ctrl, importance = TRUE)
4. Bootstrapping (Repetitive Sampling)
set.seed(2) tuned <- train(dev[, -1], dev[,1], method = "rf" , ntree =10)
Note : By default, it creates 25 repetitive samples.
If you want to set 10 repetitive samples.
tuned <- train(dev[, -1], dev[,1], method = "rf" , ntree =10 , trControl= trainControl(method="boot", number=10))
Deepanshu founded ListenData with a simple objective - Make analytics easy to understand and follow. He has over 10 years of experience in data science. During his tenure, he worked with global clients in various domains like Banking, Insurance, Private Equity, Telecom and HR.
Share Share Tweet