add monotonic constraints to workflow using tidymodels

[rdataforge]

I am working with tidymodels , so far so good. But sometimes is interesting to add monotonic constraints in the model (forcing to be always increasing, decreasing (1 , -1). This constraint is passed as a vector referring to every column in the dataframe.

Example code (non tidymodel):

# Train the model.
  xgb_model <- xgboost::xgb.train(params =  params,
                                  data=data,
                                  nrounds = num_trees,
                                  print_every_n = 5,
                                  metrics = "auc",
                                  monotone_constraints = c(1,-1,1,-1,1,-1,0,0,0,-1,1)
                                  maximize = F,
                                  tree_method = tree_method,
                                  verbose=verbose
  )

How could I insert this monotonic constraint in tidymodels?

Thanks in advance

[Max]

As long as you are not doing any filtering of predictors, you should be able to pass the vector as an engine argument:

library(parsnip)
boost_tree() %>% 
  set_mode("regression") %>% 
  set_engine("xgboost", monotone_constraints = c(1,-1,1,-1,1,-1,0,0,0,-1,1)) %>% 
  translate()
#> Boosted Tree Model Specification (regression)
#> 
#> Engine-Specific Arguments:
#>   monotone_constraints = c(1, -1, 1, -1, 1, -1, 0, 0, 0, -1, 1)
#> 
#> Computational engine: xgboost 
#> 
#> Model fit template:
#> parsnip::xgb_train(x = missing_arg(), y = missing_arg(), monotone_constraints = c(1, 
#>     -1, 1, -1, 1, -1, 0, 0, 0, -1, 1), nthread = 1, verbose = 0)

^{Created on 2022-01-03 by the reprex package (v2.0.0)}