CompositeStellib

class beast.physicsmodel.stars.stellib.CompositeStellib(osllist, *args, **kwargs)

Bases: beast.physicsmodel.stars.stellib.Stellib

Generates an object from the union of multiple individual libraries

Attributes Summary

source
wavelength	return a common wavelength sampling to all libraries.

Methods Summary

genQ(self, qname, r, \*\*kwargs)	Generate a composite value from a previously calculated
genSpectrum(self, T0[, g0, Z0, weights])	Generate a composite sprectrum
gen_spectral_grid_from_given_points(self, pts)	Reinterpolate a given stellar spectral library on to an Isochrone grid
get_boundaries(self[, dlogT, dlogg])	Returns the closed boundary polygon around the stellar library with given margins
interp(self, T0, g0, Z0, L0[, dT_max, eps, …])	Interpolation of the T,g grid
interpMany(self, T0, g0, Z0, L0[, dT_max, …])	run interp on a list of inputs and returns reduced results
which_osl(self, xypoints[, dlogT, dlogg])	Returns the library indice that contains each point in xypoints

Attributes Documentation

source

wavelength

return a common wavelength sampling to all libraries. This can be used to reinterpolate any spectrum onto a common definition

Methods Documentation

genQ(self, qname, r, **kwargs)

Generate a composite value from a previously calculated

interpolation Works on 1 desired star or a population of stars

Parameters:

qname: str

quantity name from self.grid

r: (osl, r) tuple

osl: is the library index starting from 1. 0 means no coverage. r: is the result from interp call on the corresponding library.

Returns:

q: float

value (from weighted sum)

genSpectrum(self, T0, g0=None, Z0=None, weights=None, **kwargs)

Generate a composite sprectrum

Does the interpolation or uses a previously calculated interpolation Works on 1 desired star or a population of stars

Parameters:

T0: ndarray(float)

log(Teff) to obtain

g0: ndarray(float)

log(g) to obtain

Z0: ndarray(float)

metallicity values

weights: ndarray(float)

individual weights of each star

**kwargs forwarded to interp(Many)

Returns:

s: ndarray

an array containing the composite spectrum reinterpolated onto self.wavelength

Note

if T0 and g0 are iterable, it calls interpMany

gen_spectral_grid_from_given_points(self, pts, bounds={'dlogT': 0.1, 'dlogg': 0.3})

Reinterpolate a given stellar spectral library on to an Isochrone grid

Parameters:

pts: dict like structure of points

dictionary like or named data structure of points to interpolate at pts must contain: logg surface gravity in log-scale logT log of effective temperatures (in Kelvins) logL log of luminosity in Lsun units Z metallicity

bounds: dict

sensitivity to extrapolation (see :func: Stellib.get_boundaries) default: {dlogT:0.1, dlogg:0.3}

Returns:

g: SpectralGrid

Spectral grid (in memory) containing the requested list of stars and associated spectra

get_boundaries(self, dlogT=0.1, dlogg=0.3, **kwargs)

Returns the closed boundary polygon around the stellar library with given margins

Parameters:

s: Stellib

Stellar library object

dlogT: float

margin in logT

dlogg: float

margin in logg

Returns:

b: ndarray[float, ndim=2]

(closed) boundary points: [logg, Teff] (or [Teff, logg] is swap is True)

Note

as computing the boundary could take time, it is saved in the object and only recomputed when parameters are updated

interp(self, T0, g0, Z0, L0, dT_max=0.1, eps=1e-06, bounds={})

Interpolation of the T,g grid

Interpolate on the grid and returns star indices and associated weights, and Z. 3 to 12 stars are returned. It calls _interp_, but reduce the output to the relevant stars.

Parameters:

T0: double

log(Teff) to obtain

g0: double

log(g) to obtain

T: double

log(Teff) of the grid

g: double

log(g) of the grid

dT_max: float

If, T2 (resp. T1) is too far from T compared to T1 (resp. T2), i2 (resp. i1) is not used. (see below for namings)

eps: foat

temperature sensitivity under which points are considered to have the same temperature

bounds: dict

sensitivity to extrapolation (see :func: Stellib.get_boundaries) default: {dlogT:0.1, dlogg:0.3}

Returns:

(osl, r): tuple

osl: is the library index starting from 1. 0 means no coverage. r: is the result from interp call on the corresponding library. a 3 to 12 star indexes and associated weights

interpMany(self, T0, g0, Z0, L0, dT_max=0.1, eps=1e-06, weights=None, bounds={}, pool=None, nthreads=1)

run interp on a list of inputs and returns reduced results

Interpolation of the T,g grid at Z0 metallicity

Interpolate on the grid and returns star indices and associated weights, and Z. 3 to 12 stars are returned. It calls _interp_, but reduce the output to the relevant stars.

Parameters:

T0: ndarray(float)

log(Teff) to obtain

g0: ndarray(float)

log(g) to obtain

Z0: ndarray(float)

metallicity values

L0: ndarray(float)

luminosity values

dT_max: float

If, T2 (resp. T1) is too far from T compared to T1 (resp. T2), i2 (resp. i1) is not used. (see below for namings)

eps: foat

temperature sensitivity under which points are considered to have the same temperature

weights: ndarray(float)

luminosity weigths to apply after interpolation

bounds: dict

sensitivity to extrapolation (see :func: Stellib.get_boundaries) default: {dlogT:0.1, dlogg:0.3}

pool: Pool-like object

specify a multiprocessing pool for parallel processing

nthreads: int

number of processes to use by default

Returns:

(osl, r): tuple

osl is the library index starting from 1. 0 means no coverage. r is the result from interp call on the corresponding library. a 3 to 12 star indexes and associated weights

which_osl(self, xypoints, dlogT=0.0, dlogg=0.0)

Returns the library indice that contains each point in xypoints

The decision is made from a two step search:

• first, each point is checked against the strict boundary of each library (i.e., dlogT = 0, dlogg = 0).
• second, if points are not found in strict mode, the boundary is relaxed and a new search is made.

Each point is associated to the first library matching the above

conditions.

Parameters:

xypoints: sequence

a sequence of N logg, logT pairs.

dlogT: float

margin in logT

dlogg: float

margin in logg

Returns:

res: ndarray(dtype=int)

a ndarray, 0 meaning no library covers the point, and 1, … n,

for the n-th library