This is a test version of Biostars. For the public version, visit https://www.biostars.org.
R encountering fatal error and quitting the session

Hey everyone,

I am trying to run a multiple-output Random Forest prediction using the randomForestSRC package using the rfsrc() command. I'm trying to run with ntree = 10, using my response variables [ 21 ] using a dataframe of [ 3670x217 ].

Line of code I'm trying to execute:

#training the RF model
rf_model <- rfsrc(response_formula_subset, data = train_data, na.action = "na.omit", ntree = 10)

R keeps showing this message with the symbol of a bomb saying R encountered a fatal error, needs to quite session and asking to restart the session.

I tried using a subset of the data (as you can see in the code, the subset includes only 5 responses). There would be roughly just under 200 predictors.

Really helpful if someone can explain why this is happening!

Thanks in advance!

r randomforestsrc random-forest
• 1,851 views • link
updated by Dunois • Dunois • written by anjalivasudevan98 • anjalivasudevan98

Looks like a memory related problem. Have you tried with save.memory = TRUE ?

• 0 views • link
written by manaswwm manaswwm

Can you watch memory utilisation as you run the model? Task manager on windows works.

• 0 views • link
written by dthorbur dthorbur

Have you tried running this in a fresh, clean R session with no other (unnecessary) packages and data being loaded into memory? How much system memory do you have available? How much memory is free right before rfsrc() is invoked? As others here have stated, R is probably running out of memory trying to execute the classifier.

You might also want to consider sharing some data with us (using [dput()][1]) to help us try and replicate this issue ourselves.

• 0 views • link
written by Dunois • Dunois

0 answers

No answers yet.

Log in to answer this question.

More posts like this