VanDerCorputPolicy

class probnum.quad.solvers.policies.VanDerCorputPolicy(batch_size, measure)

Bases: Policy

Pick nodes from the van der Corput sequence.

The van der Corput sequence [1] is

\[0.5, 0.25, 0.75, 0.125, 0.625, \ldots\]

If the integration domain is not [0, 1], the van der Corput sequence is linearly mapped to the domain. The domain must be finite.

Parameters:

  • batch_size (IntLike) – Size of batch of nodes when calling the policy once.

  • measure (IntegrationMeasure) – The integration measure with finite domain.

Raises:

References

Attributes Summary

requires_rng

Whether the policy requires a random number generator when called.

Methods Summary

__call__(bq_state, rng)

Find nodes according to the policy.

van_der_corput_sequence(n_start[, n_end])

Returns elements n_start, n_start + 1, ..., n_end - 1 in the van der Corput sequence.

Attributes Documentation

requires_rng

Methods Documentation

__call__(bq_state, rng)[source]

Find nodes according to the policy.

Parameters:

  • bq_state (BQState) – State of the BQ belief.

  • rng (Generator | None) – A random number generator.

Returns:

shape=(batch_size, input_dim) – Nodes found according to the policy.

Return type:

nodes

static van_der_corput_sequence(n_start, n_end=None)[source]

Returns elements n_start, n_start + 1, …, n_end - 1 in the van der Corput sequence.

\[0.5, 0.25, 0.75, 0.125, 0.625, \ldots\]

If no n_end is given, only a single element in the sequence is returned.

Parameters:

  • n_start (int) – First element of the van der Corput to be included (inclusive).

  • n_end (int | None) – Last element of the van der Corput to be included (exclusive). If not given, only the n_start element is returned.

Returns:

Array containing elements from n_start to n_end - 1 of the van der Corput sequence.

Return type:

vdc_seq