global_optimisation ============================= .. py:module:: jaxns.experimental.global_optimisation .. rubric:: :code:`jaxns.experimental.global_optimisation` .. rubric:: Module Contents .. py:class:: GlobalOptimisationState Bases: :py:obj:`NamedTuple` .. py:attribute:: key :type: jaxns.internals.types.PRNGKey .. py:attribute:: samples :type: jaxns.nested_samplers.common.types.SampleCollection .. py:attribute:: num_samples :type: jaxns.internals.types.IntArray .. py:attribute:: relative_spread :type: jaxns.internals.types.FloatArray .. py:attribute:: absolute_spread :type: jaxns.internals.types.FloatArray .. py:attribute:: num_likelihood_evaluations :type: jaxns.internals.types.IntArray .. py:class:: GlobalOptimisationResults Bases: :py:obj:`NamedTuple` .. py:attribute:: U_solution :type: jaxns.internals.types.UType .. py:attribute:: X_solution :type: jaxns.internals.types.XType .. py:attribute:: solution :type: jaxns.internals.types.LikelihoodInputType .. py:attribute:: log_L_solution :type: jaxns.internals.types.FloatArray .. py:attribute:: log_L_progress :type: jaxns.internals.types.FloatArray .. py:attribute:: num_likelihood_evaluations :type: jaxns.internals.types.IntArray .. py:attribute:: num_samples :type: jaxns.internals.types.IntArray .. py:attribute:: termination_reason :type: jaxns.internals.types.IntArray .. py:attribute:: relative_spread :type: jaxns.internals.types.FloatArray .. py:attribute:: absolute_spread :type: jaxns.internals.types.FloatArray .. py:class:: GlobalOptimisationTerminationCondition Bases: :py:obj:`NamedTuple` .. py:attribute:: max_likelihood_evaluations :type: Optional[jaxns.internals.types.IntArray] :value: None .. py:attribute:: log_likelihood_contour :type: Optional[jaxns.internals.types.FloatArray] :value: None .. py:attribute:: rtol :type: Optional[jaxns.internals.types.FloatArray] :value: None .. py:attribute:: atol :type: Optional[jaxns.internals.types.FloatArray] :value: None .. py:attribute:: min_efficiency :type: Optional[jaxns.internals.types.FloatArray] :value: None .. py:class:: SimpleGlobalOptimisation Simple global optimisation leveraging building blocks of nested sampling. .. py:attribute:: sampler :type: jaxns.samplers.abc.AbstractSampler .. py:attribute:: num_search_chains :type: int .. py:attribute:: model :type: jaxns.framework.bases.BaseAbstractModel .. py:attribute:: shell_frac :type: float :value: 0.5 .. py:attribute:: devices :type: Optional[jaxlib.xla_client.Device] :value: None .. py:attribute:: verbose :type: bool :value: False .. py:method:: __post_init__()