flexsurvreg objects are created from the flexsurv package. They differ from survreg in that the fitted models are not limited to certain parametric distributions. Users can define their own distribution, or leverage distributions like the generalized gamma, generalized F, and the Royston-Parmar spline model.
Examples
# Load libraries
library(parsnip)
library(flexsurv)
# Create model and fit
flexsurvreg_fit <- surv_reg(mode = "regression", dist = "gengamma") %>%
set_engine("flexsurv") %>%
fit(Surv(Tstart, Tstop, status) ~ trans, data = bosms3)
#> Warning: `surv_reg()` was deprecated in parsnip 0.1.6.
#> ℹ Please use `survival_reg()` instead.
out <- butcher(flexsurvreg_fit, verbose = TRUE)
#> ✖ The butchered object is 3.68 kB larger than the original. Do not butcher.
# Another flexsurvreg model object
wrapped_flexsurvreg <- function() {
some_junk_in_environment <- runif(1e6)
fit <- flexsurvreg(Surv(futime, fustat) ~ 1,
data = ovarian, dist = "weibull")
return(fit)
}
out <- butcher(wrapped_flexsurvreg(), verbose = TRUE)
#> ✖ The butchered object is 2.83 kB larger than the original. Do not butcher.