coherent_score_mz_fast.CoherentScoreMZ.comblist2cs#

Back to Class page

Summary#

Takes the return value of trigger2comblist and returns the coherent score by calling the jitted function

Signature#

def comblist2cs(self, timeseries, offsets, nfacs, gtype = 1, nsamples = 10000, **kwargs)
Input variables#

Name

Type

Default

Description

timeseries

Tuple of length n_detectors with n_samp x 3 array with Re(z), Im(z)

offsets

Array of length n_detector -1 with shifts to apply to the timeseries > H1 to bring them to zero lag

nfacs

Array of length n_detector with detector sensitivities

gtype

1

If 0, turns off the integration over distance

nsamples

10000

Number of samples to use in the monte carlo

\*\*kwargs

Generic variable to capture any extra arguments

Output variables#

Return annotation

Docstring type

Description

None

Coherent score, defined as 2 \* log (p(H_1) / p(H_0))

Docstring#

Takes the return value of trigger2comblist and returns the coherent
score by calling the jitted function
:param timeseries:
    Tuple of length n_detectors with n_samp x 3 array with Re(z), Im(z)
:param offsets:
    Array of length n_detector -1 with shifts to apply to the
    timeseries > H1 to bring them to zero lag
:param nfacs: Array of length n_detector with detector sensitivities
:param gtype: If 0, turns off the integration over distance
:param nsamples: Number of samples to use in the monte carlo
:param kwargs: Generic variable to capture any extra arguments
:return: Coherent score, defined as 2 * log (p(H_1) / p(H_0))