improve error messages #80
Conversation
| @@ -89,6 +89,7 @@ | |||
| #' cart_pca_bag | |||
| #' } | |||
| #' @export | |||
| #' @include validate.R | |||
| } | ||
|
|
||
| # TODO extend to use mode from model object(s) |
There was a problem hiding this comment.
We could eventually extend baguette to censored regression models. There is a lot of code that looks at the data to get the mode or assumes that they are either regression or classification.
| } | ||
|
|
||
| if (!is.na(msg)) { | ||
| # escape any brackets in the error message |
There was a problem hiding this comment.
A long way to get there but this should make much much nicer error messages when models fail
| Code | ||
| bagger(Sepal.Length ~ ., data = iris, times = 2L, base_model = "C5.0") | ||
| Condition | ||
| Error in `validate_outcomes_are_factors()`: |
There was a problem hiding this comment.
All of our snapshots are going to change once these hardhat functions can take calls
There was a problem hiding this comment.
Looking good!
by looking at the snapshots, i see some calls aren't passed through to hardhat::validate_outcomes_are_factors() and
validate_outcomes_are_univariate(). these don't yet have call arguments so we have to wait for {hardhat} for them
there is also a call to stop() on line 100 of R/bagger.R that needs to be converted. Wasn't able to tag directly as it is "outside of github scope"
Closes #70
This needed sooo much upkeep that it was difficult to separate the different modes of error upgrades.
This PR adds type checkers, refines cli messages, routes calls to the user-facing function, and cleans up a bunch of dplyr/purrr calls that lack namespaces.