"Error in stri_detect_regex" when using xgb.importance in R


#1

Hi everybody,

Run into a problem since I updated R libraray of xgboost to 0.82.1 (on Windows 10).
Even when I run the following example that is provided in R help, the xgb.importance does lead to an error:

data(agaricus.train, package='xgboost')
bst <- xgboost(data = agaricus.train$data, label = agaricus.train$label, max_depth = 2, 
                        eta = 1, nthread = 2, nrounds = 2, objective = "binary:logistic")
[1]	train-error:0.046522 
[2]	train-error:0.022263 
xgb.importance(model = bst)

Error in stri_detect_regex(text, "yes=(\\d+),no=(\\d+)") : 
   argument `opts_regex` should be a list

I’m not able to find out what the problem is. Any suggestions or solutions? Thank you!


#2

Ok, found the problem. There was a problem when updating the library “stringi”. After reinstalling that, everything works just fine. Sorry for the inconvenience!