public

jaxns.experimental.public

Module Contents

class DefaultGlobalOptimisation(model, num_search_chains=None, num_parallel_workers=1, s=None, k=None, gradient_slice=False)

Default global optimisation class.

A global optimisation class that uses 1-dimensional slice sampler for the sampling step and decent default values.

Parameters:

• model (jaxns.framework.bases.BaseAbstractModel) – a model to perform global optimisation on

• num_search_chains (Optional[int]) – number of search chains to use. Defaults to 20 * D.

• num_parallel_workers (int) – number of parallel workers to use. Defaults to 1. Experimental feature. If set creates a pool of identical workers and runs them in parallel.

• s (Optional[int]) – number of slices to use per dimension. Defaults to 1.

• k (Optional[int]) – number of phantom samples to use. Defaults to 0.

• gradient_slice (bool) – if true use gradient information to improve.

__call__(key, term_cond=None, finetune=False)

Runs the global optimisation.

Parameters:

• key (jaxns.internals.types.PRNGKey) – PRNGKey

• term_cond (Optional[jaxns.experimental.GlobalOptimisationTerminationCondition]) – termination condition

• finetune (bool) – whether to use gradient-based fine-tune. Default False because not all models have gradients.

Returns:

results of the global optimisation

Return type:

jaxns.experimental.GlobalOptimisationResults

summary(results, f_obj=None)

Parameters:

• results (jaxns.experimental.GlobalOptimisationResults) –

• f_obj (Optional[Union[str, TextIO]]) –