Doesn't work with latest hesim

[vladimir-spintech]

Steps to reproduce:

• Follow the tutorial

Issue:
Can't construct cost models:

> costmods <- create_costmods(n = n_samples, struct = struct, patients = pats, ae_probs = ae_probs)
Error: There must only be one row for each combination of sample, strategy_id, state_id, and time_start in 'tbl'.

Appears to be broken starting from this commit: hesim-dev/hesim@fdc41c0

[dincerti]

Yeah, this is as expected. A minor update will be needed given the change in hesim.

But it should still work with the hesim version on CRAN. I'd wait to do any updates until the development version of hesim moves to CRAN.

[munenobuk]

Still, the same error happening.

[dincerti]

Still, the same error happening.

Yes, the IVI-NSCLC package will need to be updated to reflect changes in hesim.

[munenobuk]

What modifications are needed to make it usable?

[munenobuk]

txseq_dt doesn’t fit to the tbl in the stateval_tbl, since rows are not seen as unique.
I cannot find any more clues and what to do.