16:37 Changeset [17998] by gkronber

#3127 added HEAL.VarPro for L2 regularized variable projection

15:30 Changeset [17997] by gkronber

#3127: merged r17992:17993 from trunk to branch

15:26 Changeset [17996] by gkronber

#3087 reverse merge of 17991


18:28 Changeset [17995] by dpiringe


  • added additional parameters to enable different evaluation options
  • added additive restrictions
  • added additional implementations for dynamic restrictions:
    • dynamic intervalls
    • exponatial smoothing
    • rising multiplier
  • adapted IntervalUtil to get model bounds and refactored some sections
  • adapted ShapeConstraintsParser for added features
  • added a ResultCollection in SymbolicRegressionSolution for shape constraint violations
18:16 Changeset [17994] by dpiringe


  • added MetaModelAnalyzer to analye solutions for additinal problem instances


08:23 Changeset [17993] by gkronber

#3073: changed compatibility check for IntervalInterpreter to be more specific for integer powers (see r17780)

08:01 Changeset [17992] by gkronber

#3017 minor fixes in comments


21:35 Changeset [17991] by gkronber

#3128: first dump of exploratory work-in-progress code to make sure the working copy is not lost.

21:28 Changeset [17990] by gkronber

#3128 created a branch

21:26 Ticket #3128 (Prediction intervals for non-linear regression solutions) created by gkronber
We should be able to produce prediction intervals for all symbolic …
21:15 Changeset [17989] by gkronber

#3087: updated native dlls for NativeInterpreter to a version that runs on Hive infrastructure. Some smaller changes because of deviations in the independently developed implementations (in particular enum types).

20:40 Changeset [17988] by gkronber

#3127: initial import of VarProMRGP implementation (depends on new NativeInterpreter which supports VarPro)

20:33 Changeset [17987] by gkronber

#3127 create a branch to commit my exploratory code

20:30 Ticket #3127 (Exploration of multiple regression genetic programming MRGP) created by gkronber
Not a correct implementation of MRGP …


19:36 Changeset [17986] by gkronber

#3106 Fixed a bug: training data were not resampled in restarts for local search. Made a few minor changes to match the description in the paper (introduced two parameters for local search and changed likelihood of variables to be included.)

Note: See TracTimeline for information about the timeline view.