Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

add BART survival learner #290

Merged
merged 49 commits into from
Oct 17, 2023
Merged

add BART survival learner #290

merged 49 commits into from
Oct 17, 2023

Conversation

bblodfon
Copy link
Collaborator

@bblodfon bblodfon commented Aug 17, 2023

I "hacked" some stuff in this Bayesian learner, so some things could be improved. Notes:

  • Inside predict() I need the train data as it is transformed by BART using one of their internal functions. I saved that in the self$state slot. Maybe there is a better way?
  • Since this a Bayesian model, it outputs not just a survival matrix (obs x times) but also posterior draws (3rd dimension) which is nice for uncertainty quantification. Used the self$state slot for this again. I added an example on how to use the full posterior survival matrix.
  • Added a parameter to choose the type of importance calculated, no design problems I guess with that?

@bblodfon
Copy link
Collaborator Author

@sebffischer to check/discuss during the workshop?

R/learner_BART_surv_bart.R Outdated Show resolved Hide resolved
R/learner_BART_surv_bart.R Show resolved Hide resolved
R/learner_BART_surv_bart.R Outdated Show resolved Hide resolved
R/learner_BART_surv_bart.R Outdated Show resolved Hide resolved
R/learner_BART_surv_bart.R Outdated Show resolved Hide resolved
R/learner_BART_surv_bart.R Outdated Show resolved Hide resolved
R/learner_BART_surv_bart.R Outdated Show resolved Hide resolved
R/learner_BART_surv_bart.R Outdated Show resolved Hide resolved
tests/testthat/test_BART_surv_bart.R Outdated Show resolved Hide resolved
vignettes/extending.Rmd Show resolved Hide resolved
* will work with latest version of distr6
@bblodfon
Copy link
Collaborator Author

bblodfon commented Sep 8, 2023

Depends on mlr-org/mlr3proba#330

@bblodfon
Copy link
Collaborator Author

@sebffischer we are ready to merge! The RCMD checks fail due to recent updates in aorsf, I guess @bcjaeger you plan to update the learner in another PR?

@bcjaeger
Copy link
Contributor

Sure, I will check it out now.

R/learner_BART_surv_bart.R Outdated Show resolved Hide resolved
R/learner_BART_surv_bart.R Outdated Show resolved Hide resolved
R/learner_BART_surv_bart.R Outdated Show resolved Hide resolved
R/learner_BART_surv_bart.R Outdated Show resolved Hide resolved
R/learner_BART_surv_bart.R Outdated Show resolved Hide resolved
R/learner_BART_surv_bart.R Outdated Show resolved Hide resolved
R/learner_BART_surv_bart.R Outdated Show resolved Hide resolved
R/learner_BART_surv_bart.R Outdated Show resolved Hide resolved
R/learner_BART_surv_bart.R Outdated Show resolved Hide resolved
@sebffischer sebffischer merged commit 2dbce82 into mlr-org:main Oct 17, 2023
3 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants