0

Hi every one I am working on shiny app using bnlearn for Bayesian networks and using r studio I get a fatal error and when I use R GUI I get this error 

** caught segfault ***

address 0xfffffffc0fcd6248, cause 'memory not mapped' Traceback: 1:
.Call("mappred", node = node, fitted = fitted, data = data, n =
as.integer(n), from = from, prob = prob, debug = debug) 2:
map.prediction(node = node, fitted = object, data = data, n =
extra.args$n, from = extra.args$from, prob = prob, debug = debug) 3:
predict.bn.fit(values10$fitted, input$targetvariables, combn2, prob =
TRUE, method = "bayes-lw") 4: predict(values10$fitted,
input$targetvariables, combn2, prob = TRUE, method = "bayes-lw") 5:
t(attr(predict(values10$fitted, input$targetvariables, combn2, prob =
TRUE, method = "bayes-lw"), "prob")) 6: observeEventHandler(...) 7:
..stacktraceon..(observeEventHandler(...)) 8: handlerFunc() 9:
..stacktraceon..(expr) 10: contextFunc() 11: env$runWith(self, func)
12: withReactiveDomain(.domain, { env <- .getReactiveEnvironment()
.graphEnterContext(id) on.exit(.graphExitContext(id), add = TRUE)
env$runWith(self, func)}). ......

I tried to get the latest version of R and rstudio but it still happens

is it my computer memory that is small ?? or something else

thanks in advance

neilfws
  • 32,751
  • 5
  • 50
  • 63
ou2105
  • 165
  • 3
  • 10
  • 1
    Segfaults are considered solid evidence of a bug. R and any package with compiled code are supposed to check for memory before executing operations. If you could make a reproducible example, then it should be sent to the maintainer of "bnlearn". – IRTFM Jun 15 '17 at 22:04
  • so it is not my computer that is finally to old?? – ou2105 Jun 15 '17 at 22:11
  • Hard for me to tell. My computer is 8 years old and the manufacturer considers it "obsolete". If I saw that message after rebooting and checking that I had the newest copies of R and the packages needed to run whatever code you have not displayed, I would be sending a bug report. – IRTFM Jun 15 '17 at 23:21

0 Answers0