Version 0.2.1 of the `unifiedml` R package is now available on CRAN. The key change removes the `type` parameter from `predict` in favor of `...` for greater flexibility. The post demonstrates how to wrap `ranger` and `xgboost` classifiers using S3 methods to plug them into the unified `Model` interface, covering binary classification on the Iris dataset with train/test splits and 5-fold cross-validation using `cross_val_score`.
Sort: