survxai: Visualization of the Local and Global Survival Model Explanations

Survival models may have very different structures. This package contains functions for creating a unified representation of a 'survival' models, which can be further processed by various survival explainers. Tools implemented in 'survxai' help to understand how input variables are used in the model and what impact do they have on the final model prediction. Currently, four explanation methods are implemented. We can divide them into two groups: local and global. Explanations of the methods can be found in Grudziaz et al.(2018) <doi:10.21105/joss.00961>.

Version: 0.2.2
Depends: prodlim
Imports: breakDown, ggplot2, pec, scales, survival, survminer
Suggests: CFC, covr, DALEX, knitr, markdown, randomForestSRC, rmarkdown, rms, testthat, tibble
Published: 2020-08-28
Author: Aleksandra Grudziaz [aut, cre], Alicja Gosiewska [aut], Przemyslaw Biecek [aut, ths]
Maintainer: Aleksandra Grudziaz <aleksandra.grudziaz at>
License: GPL-2 | GPL-3 [expanded from: GPL]
NeedsCompilation: no
Citation: survxai citation info
Materials: README NEWS
CRAN checks: survxai results


Reference manual: survxai.pdf
Vignettes: Custom predict function for survival models
Survival models - global explanations
How to compare models with survxai
Survival models - local explanations
Package source: survxai_0.2.2.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release: survxai_0.2.2.tgz, r-oldrel: survxai_0.2.2.tgz
Old sources: survxai archive


Please use the canonical form to link to this page.