cma.fitness_transformations

module documentation

Wrapper for objective functions like noise, rotation, gluing args

Class ComposedFunction compose an arbitrary number of functions.
Class Expensify Add a waiting time to each evaluation, to simulate "expensive" behavior
Class FBoundTransform shortcut for ComposedFunction([f, BoundTransform(bounds).transform]), see also below.
Class FixVariables Insert variables with given values,
Class Function a declarative base class, indicating that a derived class instance "is" a (fitness/objective) function.
Class GlueArguments deprecated, use functools.partial or cma.fitness_transformations.partial instead, which has the same functionality and interface.
Class IntegerMixedFunction DEPRECATED compose fitness function with some integer variables using np.floor by default.
Class IntegerMixedFunction2 compose fitness function with some integer variables using np.round by default.
Class LowEffectiveDimension Evaluate a function only in a lower dimensional subspace.
Class NeutralVariables Overwrite some variables with some given fixed values,
Class NoisyFitness apply noise via f += rel_noise(dim) * f + abs_noise(dim).
Class Rotated return a rotated version of a function for testing purpose.
Class ScaleCoordinates compose a (fitness) function with a preceding scaling and offset.
Class Shifted compose a function with a shift in x-space.
Class SomeNaNFitness transform fitness_function to return sometimes NaN
Class StackFunction a function that returns f1(x[:n1]) + f2(x[n1:]).
Variable rotate Undocumented
Function _def_abs_noise Undocumented
Function _def_rel_noise Undocumented
rotate =

Undocumented

def _def_abs_noise(dim):

Undocumented

def _def_rel_noise(dim):

Undocumented