NEWS

---

mrIML 2.2.0 (2025-11-20)

Initial bug fixes after major release of mrIML 2 to CRAN.

• mrIMLpredicts() properly handles missing X or X1.
• mrCovar() and mrVip() can handle factors without throwing a cbind error.
• Stratified test-train splits so that there is always positive and negative observation in the test set for small data sets.
• Improved internal behaviour of predict() from workflow objects.

mrIML 2.1.0 (2025-07-28)

mrIML 2.0.1

• New functionality to allow MrIML to perform joint species distribution models (JSDMs)
• New bootstrap functions and vignette
• Can also harness deep neural network models

mrIML 1.0.1

• Initial release to github
• Initial functions and vignettes