Samplers

Provide, to the RV model, new radial velocities.

The samplers are implemented with a uniform interface (this can be assumed to be true unless it is explicitly said that it is not!). Thus, if a given user input has a similar function in all samplers, it will follow a common “logic”:

• rv_step - RV_measurement :

  Most often used for numerical derivatives. The individual samplers provide more details.

• rv_window - Tuple[RV_measurement, RV_measurement]:

  Will be used to define a RV window (for each observation) inside which the sampler will have to select its next tentative RV value. This is assumed to be a “RV distance” away from the previous RV value that SBART loaded (either CCF or previous application of SBART).They must both be positive values, as we create a window of [RV_prev - rv_prior[0], RV_prev + rv_prior[1]].

chi_squared_sampler	The Chi-squared sampler implements a bounded minimization of a chi-squared curve.
Laplace_approx	Laplace's approximation to the model's posterior distribution.
MCMC_sampler	Explore the semi-Bayesian model posterior distribution with an MCMC routine (using emcee)
WindowSampler

Internals

SbartBaseSampler

Common interface of the SBART samplers.