Skip to content
Alexander Fabisch edited this page Aug 26, 2013 · 44 revisions

TODO (for release 1.1.0):

  • Document all functions / classes / ...
  • More efficient copying of training subset (use another storage order?)
  • Use base classes of vector and matrix where this is possible
  • Implement C-loss
Clone this wiki locally