sequince.steps.mutation.rejuvenation

sequince.steps.mutation.rejuvenation#

Approximate direct rejuvenation using an injected generative proposal.

Classes#

GenerativeProposal

Minimal fitted-proposal surface required for direct rejuvenation.

ProposalRejuvenation

Refit and directly redraw from an injected generative proposal.

Module Contents#

class sequince.steps.mutation.rejuvenation.GenerativeProposal[source]#

Bases: Protocol

Minimal fitted-proposal surface required for direct rejuvenation.

fit(particles, context=None, *, rng=None)[source]#
Parameters:

  • particles (Any)

  • context (Any)

  • rng (Any)

sample(n_samples, context=None, *, rng=None)[source]#
Parameters:

  • n_samples (int)

  • context (Any)

  • rng (Any)

class sequince.steps.mutation.rejuvenation.ProposalRejuvenation(target, proposal, *, transform=None)[source]#

Bases: sequince.steps.base.SMCStep

Refit and directly redraw from an injected generative proposal.

Direct proposal draws are not, in general, exact samples from the current SMC target. Unlike an MCMC mutation kernel, this step does not preserve the target exactly and has no Metropolis-Hastings correction.

A supplied transform defines the latent fitting space. Draws are mapped back to physical coordinates before the target evaluation is refreshed.

Parameters:
property proposal: GenerativeProposal#

The proposal fitted by the most recent application.

Return type:

GenerativeProposal

apply(state, context, rng)[source]#
Parameters:
Return type:

tuple[sequince.state.ParticleState, sequince.steps.context.StepContext, dict[str, Any]]

apply_functional(state, context, random_state)[source]#

Apply a functional step to particle step.

This returns the updated state, context, next random state, and any additional outputs.

This method is intended for use in functional SMC iterations where the random state is explicitly threaded through the computation.

Parameters:
Return type:

tuple[sequince.state.ParticleState, sequince.steps.context.StepContext, Any, Any]

state_dict()[source]#

Return mutable state required by subsequent SMC iterations.

Return type:

dict[str, Any]

load_state_dict(state)[source]#

Restore state produced by state_dict().

Parameters:

state (collections.abc.Mapping[str, Any])

Return type:

None