search

boost-R / mboost

Boosting algorithms for fitting generalized linear, additive and interaction models to potentially high-dimensional data. The current relase version can be found on CRAN (http://cran.r-project.org/package=mboost).
73 stars 27 forks source link

2.8.1 vs 2.9.0 differences #92

Closed seqtools closed 6 years ago

seqtools commented 6 years ago

I created a R script which works well with 2.8.1, but not 2.9.0. In the new version I got error:

Error in cvrisk.mboost(object = boost, grid = 1:mstop(boost), folds = cv, : Leave-one-out cross-validation cannot be used with ‘family = CoxPH()’

When I roll back to 2.8.1, the error disappears.

Any suggestions?

hofnerb commented 6 years ago

Unfortunately, you are not telling us what you are doing. Are you using family = CoxPH() and leave-one-out crossvalidation?

If so, that is no longer possible as the code was not working as intended. See #85

Please reopen with more details if this does not solve your issue!