recipes
recipes copied to clipboard
Pipeable steps for feature engineering and data preprocessing to prepare for modeling
**Very minor issue** `step_lag()` inherits some parameter documentation from `step_pca()`. [Here](https://github.com/tidymodels/recipes/blob/ce4f505b55f4222c58f4fbadd3d4083b61efaa14/R/lag.R#L59) it defaults the `keep_original_cols` to TRUE, whereas `step_pca()` defaults to FALSE. This causes the documentation for `step_lag()` to say...
## Feature step_impute_mode() documentation notes that, when there are multiple modes in the data, it will select a mode randomly. Even when setting a seed in your overall workflow, the...
https://github.com/tidymodels/recipes/blob/483af8a84507657d909bec152f8bb87e15fb9350/R/selections.R#L41-L46C28
See [this post](https://community.rstudio.com/t/custom-recipes-step-fails-during-bake/176576)
``` r library(tidymodels) x % mutate(y = factor(carb > 3), gear = factor(gear)) rec % add_recipe(rec) lr_wf %>% fit(data = x) #> Error in `maybe_matrix()`: #> ! Some columns are...
They are documented as `(see below)`, but then never mentioned again 😆 talk about why you might use one other the other (formula doesn't like it when there are a...
## The problem I think this might be a subtle one. If a training set: - contains a variable that is a factor - the factor knows a value as...
## The problem I'm trying to reproduce the example in the `step_novel` [function documentation](https://recipes.tidymodels.org/reference/step_novel.html#ref-examples). However when assigning a new value to the `city` predictor as per the example, instead of...
It is documented that it has been deprecated. And we stopped using it, but it is not deprecated in the lifecycle way yet.
``` \description{ These are not to be used directly by the users. \code{ellipse_check()} is deprecated. Instead, empty selections should be supported by all steps. This internal function is used for...