Note that you are strongly adviced to update your package to the most
recent version in case you have already installed it. You can do that
using the "update.packages()" function. Below you may find the most
important changes introduced by each new version of the book R
Important note: You should avoid using the versions of the
package of the branch 0.3.x. They have introduced a few modifications
on the code of the experimental comparisons that have introduced some
incompatibilities with the book code. Though the changes were
considered interesting at that time, due to these incompatibilities
I've decided to backtrack and favor the consistency with the
book. Since version 0.4.0 these changes were removed.
- VERSION 0.4.1 (2013-08-08)
- Corrected a bug on the manyNAs() function which was not working when
the second argument as an integer, as it should.
- The function SMOTE() was changed to remove its dependency on the
target variable being the last column. Now it works as expected also
if the target column is not the last column.
- Small changes in function variants() to correct a bug introduced
since version 0.3.1 of the package.
- VERSION 0.4.0 (2013-06-27)
- Functions implementing work flows that were introduced in versions
0.3.0 and 0.3.1, were eliminated to avoid incompatibilities with the
code in the book, which should be the aim of this package. In this
context, all the new functionality introduced in branch 0.3 is now
removed, and will be included in a new package specifically devoted to
the experimental comparison of predictive models that does not need to
be backward compatible with the code in the book.
- Functions slidingWindowTest() and growingWindowTest() were
re-introduced into the package.
- Functions timeseriesWF() and standardWF() were removed.
- experimentalComparison function now accepts further arguments that
are passed on to the lower level functions (e.g. crossvalidation,
- VERSION 0.2.3 (2012-03-20)
- Fixed warnings in checking on Windows versions of R 2.15.0
- VERSION 0.2.2 (2012-02-23)
- Fixed bug in function knnImputation(). Thanks to Pedro Coelho for spotting the bug.
- VERSION 0.2.1 (2011-04-14)
- Fixed a minor bug on the lofactor() function (actually on an internal function called from this one). This problem was affecting several use cases of this function, namely the code on page 203 of the book.
- VERSION 0.2.0 (2011-03-15)
- Added lofactor() function that implements the LOF algorithm by Breunig and colleagues (2000). This function is strongly based on the code previously available in package dprep by Acuna and colleagues (2009), which was removed from CRAN.
- Fixed minor bug on manyNAs() when no row satisfied the constraints satisfied by the user. In those situations the function was generating an empty index vector which could lead to undesirable side-effects, while now it generates a warning indicating an empty index.
- Modified the subset method for compExp objects. The subset method for these objects now allows for the specification of subsets of the learners, datasets and statistics through a regular expression for easier subseting on larger experimental comparisons.
- Added the utility function dsNames() to obtain the names of the data sets involved in an experimental comparison.
- Added the utility function learnerNames() to obtain the names of the learners involved in an experimental comparison.
- Added the utility function statNames() to obtain the names of the evaluation statistics involved in an experimental comparison.