Open nt-williams opened 4 years ago
Other potential errors you might want to cover:
Accidentally putting the treatment or outcome in the predictor matrix (lol) could yield an error from lmtp
instead of sl3
's non-informative one to make it easier for users to figure out what's going on.
library(lmtp)
# Code modified from Example 5.1
a <- "trt"
y <- paste0("Y.", 1:6)
cens <- paste0("C.", 0:5)
baseline <- c("W1", "W2", "trt")
progressr::with_progress({
psi5.1 <- lmtp_tmle(sim_point_surv_constant, a, y, baseline, cens = cens,
shift = static_binary_on, folds = 2,
outcome_type = "survival")
})
Error in private$.train(subsetted_task, trained_sublearners) :
All learners in stack have failed
Error in private$.train(subsetted_task, trained_sublearners) :
All learners in stack have failed
Error in self$compute_step() :
Error in private$.train(subsetted_task, trained_sublearners) :
All learners in stack have failed
Failed on Stack
Warning message:
In private$.train(subsetted_task, trained_sublearners) :
Lrnr_glm_TRUE failed with message: Error in data.table::setnames(subset, true_columns, columns): Some duplicates exist in 'old': [trt]
. It will be removed from the stack
You could also add a warning message that the k
argument will be ignored for point treatments if users input it.
Common errors/issues that aren't explicitly checked for should now be listed here:
make_learner()
instead ofmake_learner_stack()
with multiple learners.