v1.1.0

BlackBoxOptim v1.1.0

Main changes:

• Improve all examples in the examples folder.

• Expose most of optimize.py to the main API of the package.

• In optimize.py:

  • Surrogates are updated with the samples from last iteration.
    This only occurs if the surrogate can be accessed outside the function scope.
  • surrogate_optimization() and socemo() end execution the first time an iteration fails to acquire new points.
  • Fix the local acquisition function neval in SOCEMO.

• In acquisition.py:

  • Use Pymoo default termination methods instead of MaximumGenerationTermination.
  • WeightedAcquisition.argminscore() now returns -1 if there is no feasible point.
  • WeightedAcquisition.acquire() may now return less than n points.
  • ParetoFront.acquire() now can use information about the non-dominated sample points to initialize the genetic algorithm.
  • ParetoFront.acquire() only choose points that are expected to be non-dominated.

Full Changelog: v1.0.1...v1.1.0