CHANGELOG - Record of Changes for gcdnet
Please find below the list of the various changes and improvements that were part of each successive update.

1.0.2:
* [added] Now separate L2 penalty weights can be applied to each coefficient to allow
differential L2 shrinkage.

1.0.1:
* [fixed] Intercept update formula in the Fortran implementation for "sqsvm"
* [improved] The computation of L' (the first derivative of L)
* [improved] The computation of upper bound M for majorization

1.0.0:
* Initial release.

=======================================================================

Thanks to

Charles Geyer for many suggestions.

Teng Zhang for helping me get gcdnet into CRAN.

Patrick Zimmerman for suggesting improvements and finding bugs.

Qing Mai for finding bugs.