core

core#

nuance.core.separate_models(time, flux, X=None, gp=None, model=None)#

Returns a function to compute the mean, signal, and noise model of a light curve given the epoch, duration, and period of the signal model.

Parameters:

  • time (np.ndarray) – array of times

  • flux (np.ndarray) – array of fluxes

  • X (np.ndarray, optional) – linear model design matrix, by default a constant model

  • gp (tinygp.GaussianProcess, optional) – Gaussian process object, by default a very long scale exponential kernel

  • model (callable, optional) – model function with signature model(time, epoch, duration, period=None) -> Array, by default an empirical transit() model

Returns:

function that computes the mean, signal, and noise model of a light curve given the epoch, duration, and period of the signal model. Signature is: function(epoch, duration, period=None) -> (mean, signal, noise)

Return type:

callable

nuance.core.snr(time, flux, X=None, gp=None, model=None)#

Returns a function to compute the signal-to-noise ratio of a signal model given its epoch, duration, and period.

Parameters:

  • time (np.ndarray) – array of times

  • flux (np.ndarray) – array of fluxes

  • X (np.ndarray, optional) – linear model design matrix, by default a constant model

  • gp (tinygp.GaussianProcess, optional) – Gaussian process object, by default a very long scale exponential kernel

  • model (callable, optional) – model function with signature model(time, epoch, duration, period=None) -> Array, by default an empirical transit() model

Returns:

function that computes the signal-to-noise ratio of a signal model given its epoch, duration, and period. Signature is: function(epoch, duration, period=None) -> float

Return type:

callable

nuance.core.solve(time, flux, gp=None, X=None, model=None)#

Returns a function to compute the log likelihood of data assuming it is drawn from a Gaussian Process with a mean linear model.

X is a design matrix for the linear model whose last column is the searched signal model.

Parameters:

  • time (np.ndarray) – array of times

  • flux (np.ndarray) – array of fluxes

  • X (np.ndarray, optional) – linear model design matrix, by default a constant model

  • gp (tinygp.GaussianProcess, optional) – Gaussian process object, by default a very long scale exponential kernel

  • model (callable, optional) – model function with signature model(time, epoch, duration, period=None) -> Array, by default an empirical transit() model

Returns:

function that computes the log likelihood of data assuming it is drawn from a Gaussian Process with a mean linear model. Signature is: function(epoch, duration, period=None) -> (depth, depth_variance, log_likelihood)

Return type:

callable

nuance.core.transit(time, epoch, duration, period=None, c=12)#

Empirical transit model from Protopapas et al. 2005.

Parameters:

  • time (np.ndarray) – array of times

  • epoch (float) – signal epoch

  • duration (float) – signal duration

  • period (float, optional) – signal period, by default None for a non-periodic signal

  • c (int, optional) – controls the ‘roundness’ of transit shape, by default 12

Returns:

array of transit model values

Return type:

np.ndarray

nuance.core.transit_box(time, epoch, duration, period=1000000000000000.0)#

Box-shaped transit model.

Parameters:

  • time (np.ndarray) – array of times

  • epoch (float) – signal epoch

  • duration (float) – signal duration

  • period (float, optional) – signal period, by default None for a non-periodic signal

Returns:

array of transit model values

Return type:

np.ndarray

nuance.core.transit_exocomet(time, epoch, duration, period=None, n=3)#

Empirical exocomet transit model.

Parameters:

  • time (np.ndarray) – array of times

  • epoch (float) – signal epoch

  • duration (float) – signal duration

  • period (float, optional) – dummy parameter for compatibility with other models, by default None

  • n (int, optional) – TBD, by default 3

Returns:

array of transit model values

Return type:

np.ndarray

Contents