Simon P. Couch
Simon P. Couch
We attempted to transition some calls to `expect_silent` to `expect_snapshot` in conversion to testthat 3e in #105, but came across some inconsistencies in snapshots for `glm`, `lm` `ranger` tests. The...
"2 possible value include" in output of: ``` r dials::new_qual_param( type = "character", values = c("rmse", "logloss"), label = c(eval_metric = "Evaluation Metric") ) #> Evaluation Metric (qualitative) #> 2...
The `generate()` default `variables = !!response_expr(x)` trips up when `x` has yet to be passed to `specify` and errors uninformatively: ``` r library(infer) mtcars %>% generate(reps = 10, type =...
Closes #776. Worth noting that this error message has changed since the issue was filed, following up on #769. ``` r library(parsnip) bag_tree %>% set_mode("classification") #> Error in UseMethod("set_mode"): no...
Copies `knitr::combine_words` and friends to close #785. Could you confirm this does the trick for you, @mdneuzerling?
Closes #774, closes #459. Related to #411. The goal of this PR is to ensure that folks can pass arguments that live in the `param` argument to `xgb.train`. The [new...
Some adjustments to effectively enable + document the `sample_size` argument for the lightgbm engine. Will be followed up by a PR in bonsai—1/2 to close tidymodels/bonsai#30. The gist: https://github.com/tidymodels/parsnip/blob/713d617274b54a9f99c5b67c2cf52a7b0a3e431c/man/rmd/boost_tree_lightgbm.Rmd#L79-L81 Will...
We're starting to solidify the API and CRAN prep for [rstudio/bundle](https://github.com/rstudio/bundle) and are considering what integration with vetiver and tidymodels will look like. My thought with parsnip + extensions is...
See https://github.com/mine-cetinkaya-rundel/stat2-whole-game/issues/2 for more discussion. General idea: `plot.lm` ish methods, using ggplot2 fundamentals, for some of parsnip's bread and butter model specs would be nice.
This issue came up in a conversation with @mine-cetinkaya-rundel about teaching introductory stats / modeling courses using the tidymodels. I feel that, in some ways, parsnip’s guardrails re: `augment` make...