mlr-org/mlr

Provides a unified S3 interface across R's fragmented ML ecosystem, unifying classification, regression, clustering, and survival analysis with standardized resampling, hyperparameter tuning (irace, SMBO), and feature selection. Enables complex preprocessing pipelines that can be jointly optimized alongside model training, with built-in parallelization and OpenML integration for reproducible experiments. **Note: This package is retired; mlr3 is the recommended successor.**

1,679 stars. No commits in the last 6 months.