OptimisersΒΆ
Pints provides a number of optimisers, all implementing the Optimiser
interface, that can be used to find the parameters that minimise an
ErrorMeasure
or maximise a LogPDF
.
The easiest way to run an optimisation is by using the optimise()
method
or the OptimisationController
class.
- Running an optimisation
optimise()
OptimisationController
OptimisationController.evaluations()
OptimisationController.f_guessed_tracking()
OptimisationController.iterations()
OptimisationController.max_evaluations()
OptimisationController.max_iterations()
OptimisationController.max_unchanged_iterations()
OptimisationController.optimiser()
OptimisationController.parallel()
OptimisationController.run()
OptimisationController.set_callback()
OptimisationController.set_f_guessed_tracking()
OptimisationController.set_log_interval()
OptimisationController.set_log_to_file()
OptimisationController.set_log_to_screen()
OptimisationController.set_max_evaluations()
OptimisationController.set_max_iterations()
OptimisationController.set_max_unchanged_iterations()
OptimisationController.set_parallel()
OptimisationController.set_threshold()
OptimisationController.threshold()
OptimisationController.time()
Optimisation
Optimisation.evaluations()
Optimisation.f_guessed_tracking()
Optimisation.iterations()
Optimisation.max_evaluations()
Optimisation.max_iterations()
Optimisation.max_unchanged_iterations()
Optimisation.optimiser()
Optimisation.parallel()
Optimisation.run()
Optimisation.set_callback()
Optimisation.set_f_guessed_tracking()
Optimisation.set_log_interval()
Optimisation.set_log_to_file()
Optimisation.set_log_to_screen()
Optimisation.set_max_evaluations()
Optimisation.set_max_iterations()
Optimisation.set_max_unchanged_iterations()
Optimisation.set_parallel()
Optimisation.set_threshold()
Optimisation.threshold()
Optimisation.time()
- Optimiser base classes
Optimiser
Optimiser.ask()
Optimiser.f_best()
Optimiser.f_guessed()
Optimiser.fbest()
Optimiser.n_hyper_parameters()
Optimiser.name()
Optimiser.needs_sensitivities()
Optimiser.running()
Optimiser.set_hyper_parameters()
Optimiser.stop()
Optimiser.tell()
Optimiser.x_best()
Optimiser.x_guessed()
Optimiser.xbest()
PopulationBasedOptimiser
PopulationBasedOptimiser.ask()
PopulationBasedOptimiser.f_best()
PopulationBasedOptimiser.f_guessed()
PopulationBasedOptimiser.fbest()
PopulationBasedOptimiser.n_hyper_parameters()
PopulationBasedOptimiser.name()
PopulationBasedOptimiser.needs_sensitivities()
PopulationBasedOptimiser.population_size()
PopulationBasedOptimiser.running()
PopulationBasedOptimiser.set_hyper_parameters()
PopulationBasedOptimiser.set_population_size()
PopulationBasedOptimiser.stop()
PopulationBasedOptimiser.suggested_population_size()
PopulationBasedOptimiser.tell()
PopulationBasedOptimiser.x_best()
PopulationBasedOptimiser.x_guessed()
PopulationBasedOptimiser.xbest()
- Convenience methods
- Adam (adaptive moment estimation)
- Bare-bones CMA-ES
BareCMAES
BareCMAES.ask()
BareCMAES.cov()
BareCMAES.f_best()
BareCMAES.f_guessed()
BareCMAES.fbest()
BareCMAES.mean()
BareCMAES.n_hyper_parameters()
BareCMAES.name()
BareCMAES.needs_sensitivities()
BareCMAES.population_size()
BareCMAES.running()
BareCMAES.set_hyper_parameters()
BareCMAES.set_population_size()
BareCMAES.stop()
BareCMAES.suggested_population_size()
BareCMAES.tell()
BareCMAES.x_best()
BareCMAES.x_guessed()
BareCMAES.xbest()
- CMA-ES
CMAES
CMAES.ask()
CMAES.f_best()
CMAES.f_guessed()
CMAES.fbest()
CMAES.n_hyper_parameters()
CMAES.name()
CMAES.needs_sensitivities()
CMAES.population_size()
CMAES.running()
CMAES.set_hyper_parameters()
CMAES.set_population_size()
CMAES.stop()
CMAES.suggested_population_size()
CMAES.tell()
CMAES.x_best()
CMAES.x_guessed()
CMAES.xbest()
- Gradient descent (fixed learning rate)
GradientDescent
GradientDescent.ask()
GradientDescent.f_best()
GradientDescent.f_guessed()
GradientDescent.fbest()
GradientDescent.learning_rate()
GradientDescent.n_hyper_parameters()
GradientDescent.name()
GradientDescent.needs_sensitivities()
GradientDescent.running()
GradientDescent.set_hyper_parameters()
GradientDescent.set_learning_rate()
GradientDescent.stop()
GradientDescent.tell()
GradientDescent.x_best()
GradientDescent.x_guessed()
GradientDescent.xbest()
- Improved Rprop- (iRprop-)
IRPropMin
IRPropMin.ask()
IRPropMin.f_best()
IRPropMin.f_guessed()
IRPropMin.fbest()
IRPropMin.max_step_size()
IRPropMin.min_step_size()
IRPropMin.n_hyper_parameters()
IRPropMin.name()
IRPropMin.needs_sensitivities()
IRPropMin.running()
IRPropMin.set_hyper_parameters()
IRPropMin.set_max_step_size()
IRPropMin.set_min_step_size()
IRPropMin.stop()
IRPropMin.tell()
IRPropMin.x_best()
IRPropMin.x_guessed()
IRPropMin.xbest()
- Nelder-Mead
NelderMead
NelderMead.ask()
NelderMead.f_best()
NelderMead.f_guessed()
NelderMead.fbest()
NelderMead.n_hyper_parameters()
NelderMead.name()
NelderMead.needs_sensitivities()
NelderMead.running()
NelderMead.set_hyper_parameters()
NelderMead.stop()
NelderMead.tell()
NelderMead.x_best()
NelderMead.x_guessed()
NelderMead.xbest()
- PSO
PSO
PSO.ask()
PSO.f_best()
PSO.f_guessed()
PSO.fbest()
PSO.n_hyper_parameters()
PSO.name()
PSO.needs_sensitivities()
PSO.population_size()
PSO.running()
PSO.set_hyper_parameters()
PSO.set_local_global_balance()
PSO.set_population_size()
PSO.stop()
PSO.suggested_population_size()
PSO.tell()
PSO.x_best()
PSO.x_guessed()
PSO.xbest()
- SNES
SNES
SNES.ask()
SNES.f_best()
SNES.f_guessed()
SNES.fbest()
SNES.n_hyper_parameters()
SNES.name()
SNES.needs_sensitivities()
SNES.population_size()
SNES.running()
SNES.set_hyper_parameters()
SNES.set_population_size()
SNES.stop()
SNES.suggested_population_size()
SNES.tell()
SNES.x_best()
SNES.x_guessed()
SNES.xbest()
- xNES
XNES
XNES.ask()
XNES.f_best()
XNES.f_guessed()
XNES.fbest()
XNES.n_hyper_parameters()
XNES.name()
XNES.needs_sensitivities()
XNES.population_size()
XNES.running()
XNES.set_hyper_parameters()
XNES.set_population_size()
XNES.stop()
XNES.suggested_population_size()
XNES.tell()
XNES.x_best()
XNES.x_guessed()
XNES.xbest()