Error in estimating CV-TMLE treatment-specific means

[saleonard]

I'm running something similar to previously, starting simply to get it working and then want to use a dataset that I can send the job to a HPC cluster, and it worked fine the first time but not since. For each predictor the folds complete but then I get this:

Estimating CV-TMLE treatment-specific means.
Bin 1 of 3
(failed) (failed)

etc until

Error: every fold failed.

Then there are a bunch of warning messages, but I think this is likely the fundamental issues. Any thoughts?

Code:

g_lib <- c("SL.glm", "SL.mean")
Q_lib <- c("SL.glm", "SL.mean")

varimpact(Y = dat$y,
               A_names = setdiff(names(dat), c("y", "w1", "w2")),
               data = dat[, !colnames(dat) %in% "y"],
               Q.library = Q_lib,
               g.library = g_lib,
               impute = "median",
               verbose = TRUE,
                quantile_probs_factor = NULL,
               V = 2)

[ck37]

Sorry for the slow response on this Stephanie. The call itself looks reasonable - are you able to post any more details from the verbose call? Also is any parallelization involved? How does the distribution of the outcome variable look?

Thanks,
Chris

[saleonard]

Not a slow response - thanks, Chris! I just re-ran the same exact code and it ran fine. Strange as the other day I tried all sorts of things to troubleshoot and got that same error problem.
I think I'm good for now.

[ck37]

Good to hear, I will close this issue in that case. I also still need to work on better error-tracking to help narrow down these kinds of problems.