issues with dev xgboost

[nipnipj]

I'm trying to fit a model as follows:

xgboost_model <-
  boost_tree( ) %>%
  set_engine("xgboost") %>% 
  set_mode("regression")

wf_xgboost <- workflow() %>% 
  add_recipe(rec) %>% 
  add_model(xgboost_model)

trained_wf <- wf_xgboost %>% fit(train_data)

I get the following error:

Error in xgboost::xgb.DMatrix(data = x[trn_index, , drop = FALSE], missing = NA,  : 
  unused argument (info = info_list)

[EmilHvitfeldt]

Hello @nipnipj 👋

Can you please provide a minimal reprex (reproducible example)? The goal of a reprex is to make it as easy as possible for me to recreate your problem so that I can fix it: please help me help you!

If you've never heard of a reprex before, start by reading "What is a reprex", and follow the advice further down that page.

I'm not able to reproduce the error

library(tidymodels)

train_data <- mtcars

rec <- recipe(mpg ~ ., data = train_data)

xgboost_model <-
  boost_tree() %>%
  set_engine("xgboost") %>% 
  set_mode("regression")

wf_xgboost <- workflow() %>% 
  add_recipe(rec) %>% 
  add_model(xgboost_model)

trained_wf <- wf_xgboost %>% fit(train_data)

trained_wf
#> ══ Workflow [trained] ══════════════════════════════════════════════════════════
#> Preprocessor: Recipe
#> Model: boost_tree()
#> 
#> ── Preprocessor ────────────────────────────────────────────────────────────────
#> 0 Recipe Steps
#> 
#> ── Model ───────────────────────────────────────────────────────────────────────
#> ##### xgb.Booster
#> raw: 21.6 Kb 
#> call:
#>   xgboost::xgb.train(params = list(eta = 0.3, max_depth = 6, gamma = 0, 
#>     colsample_bytree = 1, colsample_bynode = 1, min_child_weight = 1, 
#>     subsample = 1), data = x$data, nrounds = 15, watchlist = x$watchlist, 
#>     verbose = 0, nthread = 1, objective = "reg:squarederror")
#> params (as set within xgb.train):
#>   eta = "0.3", max_depth = "6", gamma = "0", colsample_bytree = "1", colsample_bynode = "1", min_child_weight = "1", subsample = "1", nthread = "1", objective = "reg:squarederror", validate_parameters = "TRUE"
#> xgb.attributes:
#>   niter
#> callbacks:
#>   cb.evaluation.log()
#> # of features: 10 
#> niter: 15
#> nfeatures : 10 
#> evaluation_log:
#>      iter training_rmse
#>     <num>         <num>
#>         1    14.9313149
#>         2    10.9568064
#> ---                    
#>        14     0.5628964
#>        15     0.4603055

^{Created on 2024-03-21 with reprex v2.1.0}

[nipnipj]

The error persists for me.

library(tidymodels)

train_data <- mtcars

rec <- recipe(mpg ~ ., data = train_data)

xgboost_model <-
  boost_tree() %>%
  set_engine("xgboost") %>% 
  set_mode("regression")

wf_xgboost <- workflow() %>% 
  add_recipe(rec) %>% 
  add_model(xgboost_model)

trained_wf <- wf_xgboost %>% fit(train_data)
#> Error in xgboost::xgb.DMatrix(x, missing = NA, info = info_list): unused argument (info = info_list)

trained_wf
#> Error in eval(expr, envir, enclos): object 'trained_wf' not found

^{Created on 2024-03-22 with reprex v2.1.0}

The reason might be because I installed XGBOOST from source.

[simonpcouch]

Ah, looks like we'd expect to see a few breakages with the new xgboost: https://github.com/dmlc/xgboost/issues/9810. watchlist deprecated as an xgb.train() argument and info as a xgboost::xgb.DMatrix() argument.

We will address those breakages once xgboost moves to submit to CRAN and lets us know what breaks. As implemented, supporting both the dev and current CRAN versions would be quite gnarly.