profiley.helpers.lss

Helpers for Large-Scale Structure calculations

Module Contents

Functions

load_profiles(filename[, x, precision, force_match_all])	Load profiles created by save_profiles
power2xi(lnPgm_lnk, R)	Calculate the correlation function using the Hankel
save_profiles(file, x, y, R, profiles[, xlabel, ...])	Save 2d grid of profiles into file
xi2sigma(R, r_xi, xi, rho_m[, threads, full_output, ...])	Calculate the surface density from the correlation function
_xi2sig_single(R, ln_rxi, ln_1plusxi[, idx, verbose])	Auxiliary function to calculate the surface density from

profiley.helpers.lss.load_profiles(filename, x=None, precision=2, force_match_all=True)

Load profiles created by save_profiles

Parameters

filenamestr

name of the file to be read. To ensure consistency, this file should have been created with save_profiles; this is assumed but not checked

xiterable, optional

quantities for which the profiles are needed. Each dimension must correspond to the columns storing the variables in the file (e.g., z and logm). Each column in x must be a unique array.

precisionint or list of 2 ints, optional

number of decimal places with which to match x to the columns read from the file.

force_match_allbool, optional

whether if x is defined, all values of x must be present in the file, otherwise a ValueError is raised.

Returns

profilesndarray, shape (M,N,P)

grid of profiles

Rndarray, shape (P,)

radii at which the profiles were calculated

x1, x2ndarray, shapes (M,) and (N,), respectively

coordinates over which the grid of profiles was calculated

infotuple, len 5

contains the following information:

-name of the profile -radial units -name of x1 -name of x2 -dictionary of cosmological parameters, or None

Raises

ValueError: if x is not composed of unique arrays

Notes

• It is assumed but not check that the radial coordinates make sense. In particular, if any value in the radial coordinate is equal to -1, this function will raise an exception.

profiley.helpers.lss.power2xi(lnPgm_lnk, R)

Calculate the correlation function using the Hankel transform of the power spectrum

Taken from the KiDS-GGL pipeline, credit Andrej Dvornik.

Reference:

H. Ogata, A Numerical Integration Formula Based on the Bessel Functions, Publ. Res. Inst. Math. Sci. 41 (2005), 949-970. doi: 10.2977/prims/1145474602

Parameters

lnPgm_lnkscipy.interpolate.interp1d object (or equivalent)

function to calculate the natural log of the power spectrum given the natural log of the wavenumber

profiley.helpers.lss.save_profiles(file, x, y, R, profiles, xlabel='z', ylabel='logm', label='profile', R_units='Mpc', cosmo_params=None, verbose=True)

Save 2d grid of profiles into file

Parameters

filestr

output file name

x, yndarray, shapes (M,) and (N,)

variables over which the profiles have been calculated. Typically mass and redshift

Rndarray, shape (P,)

radial coordinates

profilesndarray, shape (M,N,P)

profiles to be saved

xlabel, ylabelstr, optional

names of the x and y coordinates

labelstr, optional

name of the profile being stored. Default ‘profile’.

R_unitsstr, optional

units of the radial coordinates, annotated in the comments

cosmodict, optional

cosmological parameters used to calculate the profiles, annotated as a comment

verbosebool, optional

verbosity. If True will only print the filename when finished.

-The first row contains to -1 entries in the first two columns, which are there only to make the contents of a file rectangular, and are ignored. Elements from the third on correspond to the radial coordinates. -Starting from the second row, the first two columns contain all combinations of the arrays x and y, and columns from the third on contain the profile for each combination. -The comments contain the units of the radial coordinates as well as the cosmological parameters used in the calculations. If cosmological parameters are not provided, the line says “# No cosmology provided”.

profiley.helpers.lss.xi2sigma(R, r_xi, xi, rho_m, threads=1, full_output=False, verbose=2)

Calculate the surface density from the correlation function

Parameters

Rnp.ndarray, shape ([M,[P,]]N)

projected distances at which to calculate the surface density, in Mpc. Can have up to two additional dimensions (typically mass and redshift)

r_xinp.ndarray, shape ([M,P,]Q)

radii at which the correlation function has been calculated. Can have up to two additional dimensions (say, mass and redshift)

xinp.ndarray, shape ([M,P,]Q)

correlation function, calculated at radii r_xi. Can have up to two additional dimensions (say, mass and redshift). Note that it is assumed that at least one of these additional dimensions is indeed present

rho_mfloat

average comoving matter density

threadsint, optional

number of threads to calculate the surface densities in parallel

full_outputbool, optional

whether to return the radial coordinates as well as the surface density (default False)

full_outputbool, optional

if True, return the radii as well as the surface density, both with the same shapes

verbose{0,1,2}

verbosity. 0 is quiet, 2 is full verbose

Returns

sigmanp.ndarray, shape ([M,[P,]]N)

surface density calculated at projected radii R

Rsigmanp.ndarray, shape ([M,[P,]]N)

radial coordinate reshaped to have the same shape as sigma. Returned only if full_output==True

Notes

• This function can take a long time to run, depending on the number of surface densities to be calculated. Be sure to store the results in a file.

• In order to make it run faster, everything will be passed to a single pool (if threads>1); if you would like to keep track of progress it is advised to call this function for pieces of your data and do the progress printing when calling this function.

profiley.helpers.lss._xi2sig_single(R, ln_rxi, ln_1plusxi, idx=None, verbose=True)

Auxiliary function to calculate the surface density from the correlation function for a single profile

The primary use of this function is as a helper for xi2sigma, which allows the calculation of multiple surface density profiles in parallel.

Parameters

Rnp.ndarray, shape (N,)

distances at which to calculate the surface density

ln_rxinp.ndarray, shape (M,)

natural log of the distances at which the correlation function has been calculated

ln_1plusxinp.ndarray, shape (M,)

natural log of (1+xi) where xi is the correlation function, calculated at distances Rxi

idxtype free, optional

pointers to the original array to map the results of the multiprocessing Pool used to loop through calls to this function

Returns

sig_singlelist, len N

integral required for surface density calculated at locations R. Note that for convenience this is a list and not a np.ndarray, and more importantly note that this is NOT the surface density, which corresponds instead to sigma=2*rho_m*R*sig_single.

idxthe argument idx is also returned only if specified

when calling this function