Resampling.h File Reference

The resamplers: function objects that draw ancestors in proportion to weights, which ParticleSet::resample() and the Resample upkeep action call. More...

#include <algorithm>
#include <cmath>
#include <concepts>
#include <cstddef>
#include <random>
#include <span>
#include <utility>
#include <vector>
#include <EmrysFilter/Core/Execution.h>
#include <EmrysFilter/Core/Macros.h>

Go to the source code of this file.

Classes
struct	Emrys::resamplers::Multinomial
	Draws each ancestor independently. More...
struct	Emrys::resamplers::Stratified
	Draws one uniform in each of M equal strata of [0, 1). More...
struct	Emrys::resamplers::Systematic
	Uses one uniform offset for M evenly spaced positions in [0, 1). More...
struct	Emrys::resamplers::Residual< Inner >
	Chooses index i floor( \( M w_i / \sum w \)) times deterministically, then draws the remainder with Inner from the fractional parts. More...

Namespaces
namespace	Emrys::resamplers
	The resamplers, each made by its factory (decisions.md [policy-factories]): multinomial(), stratified(), systematic(), and residual().

Concepts
concept	Emrys::Resampler
	Requirements on a resampler for scalar type Scalar and random generator G.
concept	Emrys::ResamplerWithScratch
	A Resampler that also takes scratch, called as r(weights, ancestors, rng, scratch) with scratch a std::span<Scalar> of at least weights.size() + ancestors.size() elements.
concept	Emrys::ResamplerUnder
	A Resampler that also runs under the execution policy P, called as r(policy, weights, ancestors, rng).

Functions
constexpr Multinomial	Emrys::resamplers::multinomial ()
constexpr Stratified	Emrys::resamplers::stratified ()
constexpr Systematic	Emrys::resamplers::systematic ()
template<typename Inner = Multinomial>
constexpr Residual< Inner >	Emrys::resamplers::residual (Inner inner={})

Detailed Description

The resamplers: function objects that draw ancestors in proportion to weights, which ParticleSet::resample() and the Resample upkeep action call.

Multinomial, stratified, systematic, and residual resampling, sequentially or under an execution policy, with the Resampler and ResamplerUnder concepts. Read ChiSquared.h next.