Raphael Sonabend-Friend

I don't disagree about touchy subject point. There's like two decades of argument in the literature about what AUC to use and when and how. And implementations across packages are...

> Another example I was thinking: what if for a measure, i.e. the C-index (surv.cindex), we wanted to have some confidence intervals or p-values for predictions made by a survival...

@mllg that was more of an example, here we mean more generally allowing outputs from functions to be stored within the object. So basically giving a `Prediction` class a field...

Heya thanks for raising the issue. To be very honest I've never had success in tuning {survivalsvm} successfully (even outside of this package). It's been buggy for ages and I'm...

Yup, buggy! I'm going to close the issue here. I don't think we should add a warning to the learner as in reality it will just perform badly and people...

With regards to Brier scores, we need to decide about how to calculate numerical estimates if the required analytical norm isn't accessible. Two solutions 1. Implement a good numerical estimator...

> Short question, since I'm not fully understanding what this design is doing: how would this be applied as the "adaptor" piece? I.e., what are arguments of constructor, what is...

This is a semantic issue that we often disagree over. But yes I mean the reduction that solves the survival task (hence "from") with classification (hence "to")

> My original question being, how precisely are you trying to reduce to classification? There doesn't seem to be just one way to translate predicted classification probabilities to a survival...

I think we should return to this once we start looking at pipelines in more detail. It may make sense to start with simpler workflows (e.g. boosting) before implementing reductions