1.9.1

• New function, bestLogConstant, that uses the same machinery to pick the best
  value of a constant to use when logging a variable, e.g. the one that makes
  the distribution look the most normal, especially useful for non-positive or
  zero-inflated data. Currently experimental.
• Taking out tests that failed due to dependent package update (does not impact default bestNormalize behavior). See (issue)[https://github.com/gmgeorg/LambertW/issues/3].

1.9.0

• Add S3 methods that helps step_orderNorm() to work with parallel processing.
• Add S3 methods that helps step_best_normalize() to work with parallel processing.
• Add a new transformation: the double reversed log (@rempsyc #18)
• Fix issues in CRAN checks

1.8.3

• updated print and term functionality to remain compatible with recipes.
• improve unit testing coverage on CI and locally

1.8.2

• improving scalability of boxcox in response to issue 10; thank you to Krzysztof Dyba (kadyb) for the suggestions.
• improved scalability of yeojohnson, thanks to Emil Hvitfeldt (EmilHvitfeldt) for his work on this problem for the recipes package here.
• updated tests to remain compatible with new recipes package (>0.1.16)

1.8.1

• update citation (new R Journal publication!)
• fix/add features to tidy method to work more generally, provide easy access to
  chosen transformations (responding to issue 9)

1.8.0

• Added packagedown website here: https://petersonr.github.io/bestNormalize
• Implemented GH actions (code coverage and R CMD check) via usethis in response to issue 7
• Improved scalability of ORQ transformation via n_logit_fit argument, with default of 10000. This should substantially decrease memory use of orderNorm while only minimally affecting the out-of-domain approximations.
• Updated documentation

1.7.0

• changed step_bestNormalize to step_best_normalize, responding to 8
• Fixed error in documentation regarding LambertW transformation types
  (thank you to Georg M. Goerg, the author of LambertW, for pointing this out).
• Add center_scale transform as default when standardize == TRUE
• Added error when trying to use repeated CV with much too small of folds
• Changed a few T and F to TRUE and FALSE
• Added documentation of how one can use scales and ggplot2 to visualize all transformations.
• Added butcher and axe functionality in order to improve scalability of step_* functions
• Improved tidy functionality with bestNormalize and step_best_normalize

1.6.1

Added additional customization potential and fixed a few bugs.

CRAN 1.5.0

• Added step_bestNormalize and step_orderNorm functions for implementation within recipes.
• Changed default to warn = FALSE when calling bestNormalize. If a transformation doesn't work,
  warnings will no longer be shown by default unless warn is set to TRUE.

CRAN 1.4.3

• Allow options to be passed through bestNormalize to specific transformation functions
• Slight bug fix to square root transformation (a = 0 by default, not .001)
• Slight bug fix in the "quiet" argument for bestNormalize with LOO
• Slight bug fix to plot.bestNormalize which was improperly labeling transformations
• exp_x having trouble with standardize option, so added option allow_exp_x to
  bestNormalize to allow a workaround, and changed it so if any infinite values
  are produced during the transformation, exp_x will not work (that way, bestNormalize
  will not include this in its results).
• Progress bar will now only displayed if quiet is FALSE and length(x) > 2000