recipes
recipes copied to clipboard
tidymodels
Pipeable steps for feature engineering and data preprocessing to prepare for modeling
I don't have a solution proposal to this one yet, so I'm just going to comment. The trained steps have a named argument in their step to pass around the...
This is a more general request for the whole package, but I will use [`step_integer()`](https://recipes.tidymodels.org/reference/step_integer.html) as an example. pkgdown prints the resulting tibble which is very nice. But I find...
A recipe step that imputes using random values of the non-missing data. The way I see it, it is on the other side of the variance/bias tradeoff compared to `step_impute_mean()`....
The errors you get if something goes wrong inside `step_pca()` are hard to read at best. Could the error be caught and displayed with a helpful message? Mention that `step_zv()`...
### Minimal, reproducible example: ```r test_data
recipes steps should only use `NULL` as an argument for: * deprecated arguments * arguments that are set at prep time this is to ensure we can safely remove the...
I'd like a signature along the lines of: ```r step_randomproj(recipe, ..., dimensions, error) ``` where the user can either specify the number of dimensions to project into or an acceptable...
There are no checks in `add_role()` right now, and the following is possible. ``` r library(recipes) recipe(mpg ~., data = mtcars) %>% add_role("mpg", new_role = "predictor") #> Recipe #> #>...
The use of `dummy` in step names have lead to some confusion, especially with the addition of `step_dummy_multi_choice()` and `step_dummy_extract()` which has `dummy` as a part of their name, while...
Closes #331. As mentioned in #287, this is important to be able to control how a recipe behaves when used within a workflow -- as far as I can tell,...
Metadata
Owner
Metadata
Pipeable steps for feature engineering and data preprocessing to prepare for modeling