Andrew Johnson
Andrew Johnson
Hmm good point, I should actually do some testing. I'll have a look and see if there's a noticeable difference
Nothing to add from my end! Let me know if there are any reverse-dependency checks or errors that I can help with
The error log is for a different package `HemingNM/ENMwizard`, I'd suggest opening a bug report there instead
Hmm I can't replicate this, the following works for me without issue: ```r modcode
@CCS-Lab would it be possible to push these changes to CRAN soon? Otherwise your package will break with the next rstan version
It looks like Julia is actually just wrapping the `sincos()` function in libm and it can just be called [directly](https://godbolt.org/z/9Mo1rhMj9): ```cpp #include double f(double X){ double s,c; sincos(X, &s, &c);...
@spinkney would you have a look at the general signature/idea here? Not 100% that is the way to go, but I'm not sure of a better alternative
> Thanks @andrjohns for tackling this! Do you happen to know any particular ranges which are problematic for the current implementation? I haven't done any proper testing at this point,...
@spinkney on a semi-related note - after being annoyed by the need to compile c++ whenever I'm validating against Boost's Math functions, I've started exposing them in an R package:...
Can these scalar/matrix/etc overloads all be condensed? They all follow a similar pattern/logic (afaict). We have `as_array_or_scalar()` as well as the vectorised `select()`, so the functions can be scalar/container agnostic...