useR! 2024

In machine learning (ML), transparency and interpretability are central to promoting trust and informed decision-making. This contribution introduces a novel R package for ML model summaries, centered on performance measures and interpretation methods. The package draws inspiration from the summary method for (additive/generalized) linear models in R which generates a table that encapsulates model performance, effect sizes and directions for individual variables, and model complexity. In our contribution, we extend this methodology to non-parametric ML models, creating a concise yet informative table that facilitates analogous conclusions. The clarity of the structured output can enhance and expedite the model selection process, making it a helpful tool for practitioners and researchers alike. Our talk presents the core functionality of the R package and addresses some implementation details, as well as potential pitfalls. With this, we hope to contribute some advancements in model transparency and comparability in the field of ML.