resampleParticles: Resample in place to avoid expensive copying of data...

In serrsBayes: Bayesian Modelling of Raman Spectroscopy

Description

Resample in place to avoid expensive copying of data structures, using a permutation of the ancestry vector.

Usage

resampleParticles(log_weights, ampMx, scaleMx, peaks, baselines, n_y, nwl)

Arguments

log_weights	logarithms of the importance weights of each particle
ampMx	npeaks x npart Matrix of amplitudes for each particle.
scaleMx	npeaks x npart Matrix of scale parameters for each particle.
peaks	nwl x npart Matrix containing the expectation of the Lorentzian mixture.
baselines	nwl x n_y x npart Array of smoothing splines.
n_y	number of observations
nwl	number of wavenumbers

Value

Vector of indices to the parents of the resampled particles.

References

Murray, L.M., Lee, A. & Jacob, P.E. (2015) "Parallel resampling in the particle filter" arXiv:1301.4019v3

See Also

residualResampling

serrsBayes documentation built on June 28, 2021, 5:14 p.m.