Hannah Frick
Hannah Frick
Getting closer @bcjaeger ! Once the parsnip PR is good to go, this one is as well 👍
#70 covers `multi_predict()` for glmnet's Cox models for the prediction types `"survival"` and `"linear_pred"`.
todo: `type = "time"` for `glmnet`
thanks @mattwarkentin ! :raised_hands:
We likely won't implement this immediately but are open to it and would welcome a PR!
The goal is to leave the model definition `boost_tree()` in parsnip untouched and only add an _engine_ in censored. Are you already familiar with the article on creating a parnsip...
Having `flexsurvspline()` models in censored would be great! Thanks for offering to contribute! Putting this in a new engine would work better than the distinction via the `dist` argument. We...
check if `floor_surv_mboost()` could use some of the functionality for cutting the KM estimates a la `survival_prob_cph()` so that we use the same logic everywhere
and survival probabilities for `bag_tree(engine = "rpart")`
I think `time` as an argument name for `predict()` works. There is no other argument to `predict()` that would refer to the event times so there isn't an opportunity for...