sequince.filtering.kernel

Transition kernels for sequential importance sampling.

Classes

ImportanceKernel

Define a particle transition and its importance-weight correction.

Module Contents

class sequince.filtering.kernel.ImportanceKernel

Bases: Protocol

Define a particle transition and its importance-weight correction.

sample(particles, context, rng)

Sample the next particle population from the supplied parents.

Parameters:

• particles (sequince.typing.Array)

• context (sequince.steps.context.StepContext)

• rng (orng.RandomGenerator)

Return type:

sequince.typing.Array

log_increment(previous_particles, particles, context)

Return the transition’s importance log-weight increment.

Parameters:

• previous_particles (sequince.typing.Array)

• particles (sequince.typing.Array)

• context (sequince.steps.context.StepContext)

Return type:

sequince.typing.Array | float