latinhypercube_driver.py

OpenMDAO design-of-experiments Driver implementing the Latin Hypercube and Optimized Latin Hypercube methods.

class openmdao.drivers.latinhypercube_driver.LatinHypercubeDriver(num_samples=1, seed=None, num_par_doe=1, load_balance=False)

Bases: openmdao.drivers.predeterminedruns_driver.PredeterminedRunsDriver

Design-of-experiments Driver implementing the Latin Hypercube method.

Args:

num_samples : int, optional

The number of samples to run. Defaults to 1.

seed : int or None, optional

Random seed. Defaults to None.

num_par_doe : int, optional

The number of DOE cases to run concurrently. Defaults to 1.

load_balance : bool, Optional

If True, use rank 0 as master and load balance cases among all of the other ranks. Defaults to False.

class openmdao.drivers.latinhypercube_driver.OptimizedLatinHypercubeDriver(num_samples=1, seed=None, population=20, generations=2, norm_method=1, num_par_doe=1, load_balance=False)

Bases: openmdao.drivers.latinhypercube_driver.LatinHypercubeDriver

Design-of-experiments Driver implementing the Morris-Mitchell method for an Optimized Latin Hypercube.

openmdao.drivers.latinhypercube_driver.debug(*arg)