With this PR, the default search method is now "forward" search for all kinds of models (previously, "L1" search was used by default if available).
The reason for this change is that in general, forward search is more favorable compared to L1 search (see section "Details" in ?varsel or ?cv_varsel).
If forward search takes too long (currently: longer than 3 minutes), a message is thrown, mentioning the runtime estimate. This is done by measuring the runtime for the intercept-only projection (in search_forward()) and then scaling this up to obtain an estimate of the runtime for the remaining search.
With this PR, the default search
methodis now"forward"search for all kinds of models (previously,"L1"search was used by default if available).The reason for this change is that in general, forward search is more favorable compared to L1 search (see section "Details" in
?varselor?cv_varsel).If forward search takes too long (currently: longer than 3 minutes), a message is thrown, mentioning the runtime estimate. This is done by measuring the runtime for the intercept-only projection (in
search_forward()) and then scaling this up to obtain an estimate of the runtime for the remaining search.