Skip to content

cvms 0.3.0

Choose a tag to compare

View all tags
@LudvigOlsen LudvigOlsen released this 29 Sep 10:32
v0.3.0
480c579

  • cross_validate_fn() is added. Cross-validate custom model functions.

  • Breaking change: In evaluate(), when type is multinomial, the output is now a single tibble. The Class Level Results are included as a nested tibble.

  • Breaking change: In baseline(), lmer models are now fitted with REML = FALSE by default.

  • Adds REML argument to baseline().

  • Bug fix: the control argument in cross_validate() was not being used. Now it is.

  • In cross_validate(), the model is no longer fitted twice when a warning is thrown during fitting.

  • Adds metrics argument to cross_validate() and validate(). Allows enabling the regular Accuracy metric
    in binomial or to disable metrics (will currently still be computed but not included in the output).

  • AICc is now computed with the MuMIn package instead of the AICcmodavg package, which
    is no longer a dependency.

  • Adds lifecycle badges to the function documentation.

Assets 2
Loading