Top | ![]() |
![]() |
![]() |
![]() |
NcmFitESMCMCWalkerAPSNcmFitESMCMCWalkerAPS — Ensemble sampler Markov Chain Monte Carlo walker - aps move. |
Functions
Properties
gdouble | rand-walk-prob | Read / Write / Construct |
gdouble | rand-walk-scale | Read / Write / Construct |
gboolean | use-interp | Read / Write / Construct |
Functions
ncm_fit_esmcmc_walker_aps_new ()
NcmFitESMCMCWalkerAPS * ncm_fit_esmcmc_walker_aps_new (guint nwalkers
,guint nparams
);
Creates a new NcmFitESMCMCWalkerAPS to be used
with nwalkers
.
ncm_fit_esmcmc_walker_aps_use_interp ()
void ncm_fit_esmcmc_walker_aps_use_interp (NcmFitESMCMCWalkerAPS *aps
,gboolean use_interp
);
Sets whether to use interpolation of the posterior approximation (use_interp
== TRUE)
or kernel density estimate (use_interp
== FALSE).
ncm_fit_esmcmc_walker_aps_set_rand_walk_prob ()
void ncm_fit_esmcmc_walker_aps_set_rand_walk_prob (NcmFitESMCMCWalkerAPS *aps
,const gdouble prob
);
Sets the probability of stepping using a random walk instead of sampling from the posterior approximation.
ncm_fit_esmcmc_walker_aps_set_rand_walk_scale ()
void ncm_fit_esmcmc_walker_aps_set_rand_walk_scale (NcmFitESMCMCWalkerAPS *aps
,const gdouble scale
);
Sets the scale of the random walk step.
ncm_fit_esmcmc_walker_aps_get_rand_walk_prob ()
gdouble
ncm_fit_esmcmc_walker_aps_get_rand_walk_prob
(NcmFitESMCMCWalkerAPS *aps
);
Gets current probability of stepping using a random walk instead of sampling from the posterior approximation.
ncm_fit_esmcmc_walker_aps_get_rand_walk_scale ()
gdouble
ncm_fit_esmcmc_walker_aps_get_rand_walk_scale
(NcmFitESMCMCWalkerAPS *aps
);
Gets current probability of stepping using a random walk instead of sampling from the posterior approximation.
Property Details
The “rand-walk-prob”
property
“rand-walk-prob” gdouble
The probability of making a random walk step.
Flags: Read / Write / Construct
Allowed values: [0,1]
Default value: 0.01
The “rand-walk-scale”
property
“rand-walk-scale” gdouble
The probability of making a random walk step.
Flags: Read / Write / Construct
Allowed values: >= 0
Default value: 0.5
The “use-interp”
property
“use-interp” gboolean
Whether to use interpolation to build the posterior approximation.
Flags: Read / Write / Construct
Default value: TRUE