Generates plots for mlr3proba::PredictionSurv, depending on argument type:

• "calib" (default): Calibration plot comparing the average predicted survival distribution to a Kaplan-Meier prediction, this is not a comparison of a stratified crank or lp prediction. object must have distr prediction. geom_line() is used for comparison split between the prediction (Pred) and Kaplan-Meier estimate (KM). In addition labels are added for the x (T) and y (S(T)) axes.

• "dcalib": Distribution calibration plot. A model is D-calibrated if X% of deaths occur before the X/100 quantile of the predicted distribution, e.g. if 50% of observations die before their predicted median survival time. A model is D-calibrated if the resulting plot lies on x = y.

• "preds": Matplots the survival curves for all predictions

## Usage

# S3 method for PredictionSurv
autoplot(
object,
type = "dcalib",
row_ids = NULL,
times = NULL,
xyline = TRUE,
cuts = 11L,
...
)

## Arguments

object
type

(character(1)):
Type of the plot. See description.

If type = "calib" then task is passed to $predict in the Kaplan-Meier learner. row_ids (integer()) If type = "calib" then row_ids is passed to$predict in the Kaplan-Meier learner.

times

(numeric())
If type = "calib" then times is the values on the x-axis to plot over, if NULL uses all times from task.

xyline

(logical(1))
If TRUE (default) plots the x-y line for type = "dcalib".

cuts

(integer(1))
Number of cuts in (0,1) to plot dcalib over, default is 11.

...

## Theme

The theme_mlr3() and viridis color maps are applied by default to all autoplot() methods. To change this behavior set options(mlr3.theme = FALSE).

## References

Haider H, Hoehn B, Davis S, Greiner R (2020). “Effective Ways to Build and Evaluate Individual Survival Distributions.” Journal of Machine Learning Research, 21(85), 1-63. https://jmlr.org/papers/v21/18-772.html.

## Examples

learn = lrn("surv.coxph")
p = learn$train(task, row_ids = 1:300)$predict(task, row_ids = 301:400)