recipes
recipes copied to clipboard
tidymodels
Pipeable steps for feature engineering and data preprocessing to prepare for modeling
## The problem There seems to be an issue with the `step_interact()` function when attempting to create multiple interactions and two of the variable names have more than a certain...
Right now we have a lot of different approaches going on. It would be nice if we could uniform it it a bit.
some functions let you pass in a a function as a argument (`step_dummy()`), others forces a specific naming structure (`step_bs()`)
Ref: https://stackoverflow.com/questions/76145016/step-poly-error-in-tidymodels-degree-must-be-less-than-number-of-unique-poin More of a general problem.
Name is still up for grabs. But the main idea is to let `step_select()` only affect predictors rather than everything. With the option to affect everything
I'm thinking specifically for `bake()`ing. We already ran into this a couple of times for some steps. We should add to the base testing suite It will take forever to...
This should shortcut its ability to rename, and be more inline with other `check_*()` functions
This is a very unlikely thing to happen naturally, but I'm still posting as I may discover other weird things when investigating ```r library(recipes) exp_data prep() |> bake(new_data = NULL)...
## The Problem I trained a recipe to be have raw polynomials, but when I print it out, it says it is orthogonal: ## Reproducible example ```r library(tidymodels) library(tidyverse) data(biomass,...
# Tidyup: variable input/output information in {recipes} **Champion**: Emil **Co-Champion**: Max **Status**: Draft ## Abstract The [recipes](https://recipes.tidymodels.org/) provide a pipe-able and flexible way of processing data. Each operation is done...
Metadata
Owner
Metadata
Pipeable steps for feature engineering and data preprocessing to prepare for modeling