Closed mb706 closed 1 year ago
part from the current version of the book:
We also set a featureless baseline as a fallback learner (@sec-fallback) and set "try" as our encapsulation method, which logs errors/warnings to an external file that can be read by batchtools (we will return to this in @sec-batchtools-monitoring). # featureless baseline lrn_baseline = lrn("classif.featureless", id = "featureless") # logistic regression pipeline lrn_lr = lrn("classif.log_reg") lrn_lr = as_learner(ppl("robustify", learner = lrn_lr) %>>% lrn_lr) lrn_lr$id = "logreg" lrn_lr$fallback = lrn_baseline lrn_lr$encapsulate = c(train = "try", predict = "try")
We also set a featureless baseline as a fallback learner (@sec-fallback) and set "try" as our encapsulation method, which logs errors/warnings to an external file that can be read by batchtools (we will return to this in @sec-batchtools-monitoring).
"try"
batchtools
# featureless baseline lrn_baseline = lrn("classif.featureless", id = "featureless") # logistic regression pipeline lrn_lr = lrn("classif.log_reg") lrn_lr = as_learner(ppl("robustify", learner = lrn_lr) %>>% lrn_lr) lrn_lr$id = "logreg" lrn_lr$fallback = lrn_baseline lrn_lr$encapsulate = c(train = "try", predict = "try")
Documented in the last release.
part from the current version of the book: