I want to use modelStudio and for that I need to make explainer using DALEX::explain() that takes model object as an argument. I should, but I am not sure how to provide the entire reproducible example code with data here!
When I use the following code
xgb <- boost_tree(mode = "classification",
trees = 100,
mtry = 0.7,
learn_rate = 0.15,
tree_depth = 10,
sample_size = 1) %>%
set_engine("xgboost") %>%
fit(Y ~ ., data = train)
the following explainer works:
explainer <- explain(xgb,
data = test,
y = test$Y,
predict_function = ,
label = "xgb")
but when I use the entire workflow and try extracting the model using pull_workflow_fit, it breaks saying it can't predict.
xgb1 <- final_xgb %>%
fit(data = train) %>%
pull_workflow_fit()
classes in both the cases are _xgb.Booster model_fit.
Which function in tidymodels or workflow will render the object that exactly matches with the xgb (first code chunk above) model?
