Changes in version 1.0.4 (2025-04-12)                  

Minor improvements

  - Better code formatting.
  - More return() statements.
  - Speed-up for some regression metrics.
  - Push code coverage version.

                 Changes in version 1.0.3 (2023-11-16)                  

Improvements

  - logLoss() and deviance_poisson() are now less picky about
    predictions of 0 (or 1).

                 Changes in version 1.0.2 (2023-06-05)                  

Documentation

  - Fix Latex problem in MacOS help files.
  - Slight improvements in docu.

                 Changes in version 1.0.1 (2023-05-11)                  

Documentation

  - Much more compact help files
  - Using Latex in help files

                 Changes in version 1.0.0 (2023-04-16)                  

This release bumps the package to stable version 1.0.0.

Maintenance

  - Reorganization of code
  - Switch to package::function() notation

                 Changes in version 0.5.5 (2023-03-24)                  

Maintenance

  - Use github workflows
  - Add github pages
  - Better README

Dependencies

  - Removed {dplyr} from suggested packages.

                 Changes in version 0.5.4 (2022-01-23)                  

This is a maintainance update without any code change.

  - Fixed problematic unit tests.
  - Internal restructuring how package content is being generated.

                 Changes in version 0.5.3 (2021-06-06)                  

  - Added unit tests
  - Added option reference_mean to r_squared() functions. This allows
    clean out-of-sample applications.
  - Added Murphy diagrams

                 Changes in version 0.5.2 (2021-04-16)                  

Maintainance release. rmarkdown is now an explicit "Suggested" package.

                 Changes in version 0.5.1 (2020-04-18)                  

New function

  - weighted_cor to calculate weighted correlation between actual and
    observed values.

  - prop_within to calculate weighted proportion of predictions within a
    tolerance around actual values.

Requirements

  - Reduced minimal R version from 3.5 to 3.1.

Documentation

  - Clarified that R-squared is calculated with respect to the null
    model calculated from the actual values.

                 Changes in version 0.5.0 (2020-01-08)                  

New functions

  - Elementary scoring functions for expectiles and quantiles.

  - multi_metric: A way to create a named list of performance measures
    parametrized by a parameter.

                 Changes in version 0.4.0 (2019-12-02)                  

New functions

Added the following convenience wrappers to r_squared.

  - r_squared_poisson

  - r_squared_gamma

  - r_squared_bernoulli

                 Changes in version 0.3.0 (2019-11-08)                  

New function

Added function weighted_var to calculate variance weighted by sampling
weights.

                 Changes in version 0.2.0 (2019-08-19)                  

  - Improvement of documentation and examples.

  - Better handling of Tweedie special cases.

  - More strict error handling.

  - Added median absolute error (and weighted_median, weighted_quantile)

                 Changes in version 0.1.0 (2019-07-28)                  

Initial release.