def _model_to_settings(X, in_grid_config, in_regularization_config,
in_solver_config, in_final_solver_config,
in_adaptivity_config):
grid_config = RegularGridConfiguration()
grid_config.dim_ = X.shape[1]
grid_config.level_ = in_grid_config.level
grid_config.type_ = in_grid_config.type #6 = ModLinear
grid_config.t_ = in_grid_config.T
adaptivity_config = AdpativityConfiguration()
adaptivity_config.numRefinements_ = in_adaptivity_config.num_refinements
if adaptivity_config.numRefinements_ == 0:
adaptivity_config.noPoints_ = 0
adaptivity_config.percent_ = 0
adaptivity_config.threshold_ = 0
else:
adaptivity_config.noPoints_ = in_adaptivity_config.no_points
adaptivity_config.percent_ = in_adaptivity_config.percent
adaptivity_config.threshold_ = in_adaptivity_config.treshold
solver_config = SLESolverConfiguration()
solver_config.type_ = in_solver_config.type
solver_config.maxIterations_ = in_solver_config.max_iterations
solver_config.eps_ = in_solver_config.epsilon
solver_config.threshold_ = in_solver_config.threshold
final_solver_config = SLESolverConfiguration()
final_solver_config.type_ = in_final_solver_config.type
final_solver_config.maxIterations_ = in_final_solver_config.max_iterations
final_solver_config.eps_ = in_final_solver_config.epsilon
final_solver_config.threshold_ = in_solver_config.threshold
regularization_config = RegularizationConfiguration()
regularization_config.exponentBase_ = in_regularization_config.exponent_base
regularization_config.regType_ = in_regularization_config.type
regularization_config.lambda_ = in_regularization_config.lambda_reg
regularization_config.l1Ratio_ = in_regularization_config.l1_ratio
return grid_config, adaptivity_config, solver_config, final_solver_config, regularization_config
numDims = 6 level = 4 refnums = 0 consistentGrid = False # candidateSearchAlgorithm = MakePositiveCandidateSearchAlgorithm_FullGrid # candidateSearchAlgorithm = MakePositiveCandidateSearchAlgorithm_Intersections candidateSearchAlgorithm = MakePositiveCandidateSearchAlgorithm_IntersectionsJoin # candidateSearchAlgorithm = MakePositiveCandidateSearchAlgorithm_HybridFullIntersections # interpolationAlgorithm = MakePositiveInterpolationAlgorithm_InterpolateBoundaries1d interpolationAlgorithm = MakePositiveInterpolationAlgorithm_SetToZero plot = True verbose = True code = "c++" gridConfig.dim_ = numDims gridConfig.level_ = level gridConfig.maxDegree_ = level + 1 mu = np.ones(numDims) * 0.5 cov = np.diag(np.ones(numDims) * 0.1 / 10.) dist = J([Normal(0.5, 1. / 16., 0, 1)] * numDims) # dist = MultivariateNormal(mu, cov, 0, 1) # problems in 3d/l2 # dist = J([Beta(5, 4, 0, 1)] * numDims) # problems in 5d/l3 # dist = J([Lognormal(0.2, 0.7, 0, 1)] * numDims) # problems in 5d/l3 trainSamples = dist.rvs(1000) testSamples = dist.rvs(1000) # plot analytic density if numDims == 2 and plot: