navigation links

utils.vegas_integrals module

class VegasSampler(*args, **kwargs)

Bases: utils.integral_validation.Sampler

Parameters

• integrator –

• train –

• n_survey_steps –

• n_batch –

sample(f, n_batch=10000, *args, **kwargs)

Parameters

• f (utils.integrands.Integrand) –

• n_batch (int) –

Returns

• tuple of torch.Tensor

• x,px,fx (points, pdfs, function values)

sample_non_stratified(f, n_batch=10000, *args, **kwargs)

Parameters

• f (batch callable) – function mapping numpy arrays to numpy arrays

• n_batch –

• args –

• kwargs –

sample_stratified(f, n_batch)

train_integrator(n_survey_steps, n_batch)

Train the integrator before sampling

Parameters

• n_survey_steps (int) – if train is True, how many survey steps to use for training

• n_batch (int) – maximum number of function evaluations per survey step

evaluate_integral_vegas(f, integrator, n_batch=10000, train=True, n_survey_steps=10, n_batch_survey=10000, stratified=False)

Validate a known integral using a VEGAS integrator as a sampler

Parameters

• f (utils.integrands.KnownIntegrand) –

• integrator (zunis.integration.adaptive_survey_integrator.AdaptiveSurveyIntegrator) –

• n_batch (int) –

• train (bool) – whether to train the integrator using integrator.survey

• n_survey_steps (int or None) – positional integrator.survey argument

• n_batch_survey (int or None) –

• stratified (bool) – whether to use VEGAS stratified sampling

Returns

Return type

utils.record.EvaluationRecord