Draft Surv to classif pipeline #194

mlr-org / mlr3proba

Probabilistic Learning for mlr3

https://mlr3proba.mlr-org.com/

GNU Lesser General Public License v3.0

130 stars 20 forks source link

Closed studener closed 4 months ago

bblodfon commented 5 months ago

[x] run document() to generate the docs and get the exported objects
[x] add tests
[x] make pipeline_survtoclassif (see pipeline_survtoregr) that uses the two PipeOps consesutively
[x] pammtools::as_ped => to base R => we import pammtools for now, no need to do anything further here
[x] add cut hyperparameter

adibender commented 5 months ago

In the end run some sanity check:

use task generator to generate survival data
use discSurv package + logistic regression -> compare to our pipeline (e.g. same coefficients, etc.)
use misscpecified model (without covariate that has effect) and correctly specified model (with covariate with effect) to check that c-index, brier-score etc. have better values for correctly specified model