def check_Bessel_spectrum(l, k=None, r0=100):
# Initialise survey and spectra
cosmo = cosmicpy.cosmology()
surv = cosmicpy.survey(nzparams={
'type': 'gaussian',
'cosmology': cosmo,
'r0': r0
},
zmax=5,
fsky=1.0)
spectra = cosmicpy.spectra(cosmo, surv)
survey_volume = pi**(3.0 / 2.0) * r0**3
if k is None:
k = np.logspace(log(0.0001) / log(10),
log(0.25) / log(10),
512,
endpoint=False)
kw, w = spectra.W(l, k, evol=False)
kw = kw.squeeze()
w = w.squeeze()
w_exact = np.zeros_like(w)
for k1 in range(len(k)):
w_exact[k1, :] = gaussian_bessel_window(l, r0, k[k1],
kw[k1, :]) / survey_volume
pk = cosmo.pk_lin(kw)
integrand = np.square(kw * w_exact) * pk
res = (2.0 / pi)**2 * trapz(integrand, x=kw)
cl = spectra.cl_sfb(l, k)
return k, cl, res
def check_Bessel_spectrum(l, k=None, r0=100):
# Initialise survey and spectra
cosmo = cosmicpy.cosmology()
surv = cosmicpy.survey(nzparams={'type': 'gaussian',
'cosmology': cosmo,
'r0': r0},
zmax=5,
fsky=1.0)
spectra = cosmicpy.spectra(cosmo, surv)
survey_volume = pi**(3.0 / 2.0) * r0**3
if k is None:
k = np.logspace(log(0.0001) / log(10), log(0.25) / log(10), 512,
endpoint=False)
kw, w = spectra.W(l, k, evol=False)
kw = kw.squeeze()
w = w.squeeze()
w_exact = np.zeros_like(w)
for k1 in range(len(k)):
w_exact[k1, :] = gaussian_bessel_window(l,r0, k[k1], kw[k1,:])/ survey_volume
pk = cosmo.pk_lin(kw)
integrand = np.square(kw * w_exact) * pk
res = (2.0 / pi)**2 * trapz(integrand, x=kw)
cl = spectra.cl_sfb(l, k)
return k, cl, res
def check_Bessel_window(l, k=None, r0=100):
r""" Checks the computation of the bessel window compared to the exact
solution
for Gaussian selection functions.
Parameters
----------
l : int
Order of the multipole.
r0 : real, optional
Radius parameter for the survey Gaussian selection function in Mpc/h
Returns
-------
(k, W, Wexact) : tuple of ndarrays
Window compute the discrete spherical bessel transform and the exact
window computed from theory
"""
# Initialise survey and spectra
cosmo = cosmicpy.cosmology()
surv = cosmicpy.survey(nzparams={
'type': 'gaussian',
'cosmology': cosmo,
'r0': r0
},
zmax=0.5,
fsky=1.0)
spectra = cosmicpy.spectra(cosmo, surv)
survey_volume = pi**(3.0 / 2.0) * r0**3
if k is None:
k = np.logspace(log(0.001) / log(10),
log(0.20) / log(10),
512,
endpoint=False)
W = spectra.W(l, k, k, evol=False, kmax=0.5)
W = W.squeeze()
Wexact = np.zeros_like(W)
for k1 in range(len(k)):
Wexact[k1, :] = (gaussian_bessel_window(l, r0, k[k1], k) /
survey_volume)
print("Maximum absolute error vs maximum value: " +
str(abs(W - Wexact).max()) + " " + str(Wexact.max()))
return (k, W, Wexact)
def check_Bessel_window(l, k=None, r0=100):
r""" Checks the computation of the bessel window compared to the exact
solution
for Gaussian selection functions.
Parameters
----------
l : int
Order of the multipole.
r0 : real, optional
Radius parameter for the survey Gaussian selection function in Mpc/h
Returns
-------
(k, W, Wexact) : tuple of ndarrays
Window compute the discrete spherical bessel transform and the exact
window computed from theory
"""
# Initialise survey and spectra
cosmo = cosmicpy.cosmology()
surv = cosmicpy.survey(nzparams={'type': 'gaussian',
'cosmology': cosmo,
'r0': r0},
zmax=0.5,
fsky=1.0)
spectra = cosmicpy.spectra(cosmo, surv)
survey_volume = pi**(3.0 / 2.0) * r0**3
if k is None:
k = np.logspace(log(0.001) / log(10), log(0.20) / log(10), 512,
endpoint=False)
W = spectra.W(l, k, k, evol=False, kmax=0.5)
W = W.squeeze()
Wexact = np.zeros_like(W)
for k1 in range(len(k)):
Wexact[k1, :] = (gaussian_bessel_window(l, r0, k[k1], k) /
survey_volume)
print("Maximum absolute error vs maximum value: " +
str(abs(W - Wexact).max()) +
" " +
str(Wexact.max()))
return (k, W, Wexact)
