Xavier Robin

Indeed after increasing the timeout I no longer get these empty error messages. It would be nice if the failure message included information about the timeout. Also it makes this...

That's a tough one. My gut feeling is that square-root specificity / sensitivity should go to the coords function as coordinates. But it wouldn't give you an AUC. So we...

Looks good, thanks!

Sorry for the late answer here, this week's been crazy. I guess I should stop testing deprecated stuff. I see the following errors: ```══ Failed tests ════════════════════════════════════════════════════════════════ ── Failure ('test-cov.R:76:2'):...

Deprecation warnings are gone now, but new errors popped up R-devel (a note about something in the documentation) and with R-devel on AppVeyor (Graphics version mismatch). I will try to...

The Graphics API version mismatch was occurring because of a mismatch in the ggroc version (it must have been built on a previous version of R). I found how how...

The last remaining deprecation warnings have been removed as well.

Of course the last commit broke everything for no reason. Except for this PR, inexplicably. Sorry about that 😔

I can see slow tests are being skipped. Basically all the tests of bootstrapping are skipped with skip_slow(). I tried to fix that earlier, clearly was not a success. Will...

CI is fixed now. The difference is likely coming from different dimensions between laply and vapply: ``` # old Browse[1]> dim(laply(1:boot.n, stratified.bootstrap.test, roc1=roc1, roc2=roc2, test="boot", x=NULL, paired=TRUE, auc1skeleton=auc1skeleton, auc2skeleton=auc2skeleton, .progress=progress,...