step_profile for categorical pdp

[rdataforge]

following code here:

How to use `pdp:: partial()` in `tidymodels`

This works for me but is not extensively tested. I'm not sure why we get more lines in the 2nd and 3rd plot. library(tidymodels) library(pdp) ## ----------------------------------------------------------------------------- # results should look like: lm(mpg ~ wt + I(1/disp^2), data = mtcars) %>% partial(pred.var = "disp") %>% plotPartial() ## ----------------------------------------------------------------------------- # From ?partial: # `pred.fun`: Optional prediction function that requires two arguments: # `object` and `newdata`. If specified, then the function must # return a single prediction or a vector of predictions (i.e., not # a matrix …

slightly modified code :

grid <- 
  recipe(target ~ ., data = df) %>% 
  step_profile(all_predictors(), -target, profile = vars(my_var)) %>% 
  prep() %>% 

predict(workflow_fit, grid) %>% 
  bind_cols(grid %>% select(my_var)) %>% 
  ggplot(aes(y = .pred, x = my_var)) + 
  geom_bar()

Worked flawlessly for numerical variables.
I have tried to apply the same procedure for categorical variables but all bars have the same lenght.

any hint on getting pdp plots for categorical variables using tidymodels?