Merge branch 'master' of bitbucket.org:glavaux/cosmotool

This commit is contained in:
Guilhem Lavaux 2014-07-05 22:05:27 +02:00
commit 15e824a097
8 changed files with 95 additions and 54 deletions

View File

@ -1,3 +1,5 @@
import matplotlib
matplotlib.use('Agg')
import healpy as hp
import numpy as np
import cosmotool as ct
@ -21,7 +23,8 @@ parser.add_argument('--iid', type=int, default=0)
parser.add_argument('--proj_cat', type=bool, default=False)
args = parser.parse_args()
INDATA="/nethome/lavaux/Copy/PlusSimulation/BORG/Input_Data/2m++.npy"
#INDATA="/nethome/lavaux/Copy/PlusSimulation/BORG/Input_Data/2m++.npy"
INDATA="2m++.npy"
tmpp = np.load(INDATA)
L = args.boxsize

View File

@ -3,6 +3,8 @@ import numpy as np
import cosmotool as ct
import argparse
import h5py as h5
import matplotlib
matplotlib.use('Agg')
from matplotlib import pyplot as plt
import ksz
from ksz.constants import *
@ -54,48 +56,60 @@ def build_unit_vectors(N):
return ux,uy,uz
def build_radial_v(v):
N = v[0].shape[0]
u = build_unit_vectors(N)
vr = v[0] * u[0]
vr = v[0] * u[2]
vr += v[1] * u[1]
vr += v[2] * u[2]
return vr
vr += v[2] * u[0]
def generate_from_catalog(vfield,Boxsize):
return vr.transpose()
def generate_from_catalog(vfield,Boxsize,dmin,dmax):
import progressbar as pbar
cat = np.load("2m++.npy")
profiler = KSZ_Isothermal(2.37)
cat['distance'] = cat['best_velcmb']
cat = cat[np.where((cat['distance']>100*dmin)*(cat['distance']<dmax*100))]
deg2rad = np.pi/180
Npix = 12*Nside**2
xp,yp,zp = hp.pix2vec(Nside, np.arange(Npix))
N2 = np.sqrt(xp**2+yp**2+zp**2)
ksz_template = np.zeros(12*Nside**2, dtype=np.float64)
ksz_mask = np.zeros(12**Nside**2, dtype=np.uint8)
ksz_template = np.zeros(Npix, dtype=np.float64)
ksz_mask = np.zeros(Npix, dtype=np.uint8)
pb = pbar.ProgressBar(maxval = cat.size, widgets=[pb.Bar(), pb.ETA()]).start()
pb = pbar.ProgressBar(maxval = cat.size, widgets=[pbar.Bar(), pbar.ETA()]).start()
for k,i in np.ndenumerate(cat):
pb.update(k)
pb.update(k[0])
l,b=i['gal_long'],i['gal_lat']
ra,dec=i['ra'],i['dec']
l *= deg2rad
b *= deg2rad
x0 = np.cos(l)*np.cos(b)
y0 = np.sin(l)*np.cos(b)
z0 = np.sin(b)
ra *= deg2rad
dec *= deg2rad
# x0 = np.cos(l)*np.cos(b)
# y0 = np.sin(l)*np.cos(b)
# z0 = np.sin(b)
x0 = xra = np.cos(ra)*np.cos(dec)
y0 = yra = np.sin(ra)*np.cos(dec)
z0 = zra = np.sin(dec)
DA =i['distance']/100
Lgal = DA**2*10**(0.4*(tmpp_cat['Msun']-i['K2MRS']+25))
idx0 = hp.query_disc(Nside, (x0,y0,z0), 3*rGalaxy/DA)
profiler = ksz.KSZ_Isothermal(Lgal, 2.37)
vr = interp3d(DA * x0, DA * y0, DA * z0, vfield, Boxsize)
idx0 = hp.query_disc(Nside, (x0,y0,z0), 3*profiler.rGalaxy/DA)
vr = interp3d(DA * xra, DA * yra, DA * zra, vfield, Boxsize)
xp1 = xp[idx0]
yp1 = yp[idx0]
@ -119,18 +133,33 @@ for i in xrange(args.start,args.end,args.step):
ff=plt.figure(1)
plt.clf()
v=[]
with h5.File(args.base_h5 % i, mode="r") as f:
p = wrapper_impulsion(f)
v.append(p[i] / f['density'][:])
fname = args.base_h5 % i
if False:
print("Opening %s..." % fname)
with h5.File(fname, mode="r") as f:
p = wrapper_impulsion(f)
for j in xrange(3):
v.append(p[j] / f['density'][:])
print("Building radial velocities...")
# Rescale by Omega_b / Omega_m
vr = build_radial_v(v)
vr *= -ksz_normalization*rho_mean_matter*1e6
del v
vr = build_radial_v(v)
# _,_,vr = build_unit_vectors(128)
# vr *= 1000 * 500
vr *= ksz_normalization*1e6
del v
# np.save("vr.npy", vr)
else:
print("Loading vrs...")
vr = np.load("vr.npy")
proj = generate_from_catalog(vfield,Boxsize)
print("Generating map...")
proj,mask = generate_from_catalog(vr,args.boxsize,args.depth_min,args.depth_max)
hp.write_map(args.ksz_map % i, proj)
hp.write_map((args.ksz_map % i) + "_mask", mask)
hp.mollview(proj, fig=1, coord='CG', cmap=plt.cm.coolwarm, title='Sample %d' % i, min=args.minval,
max=args.maxval)

View File

@ -53,7 +53,7 @@ def compute_ref_power(L, N, cosmo, bins=10, range=(0,1), func='HU_WIGGLES'):
return bin_power(p.compute(k)*cosmo['h']**3, L, bins=bins, range=range)
def run_generation(input_borg, a_borg, a_ic, cosmo, supersample=1, do_lpt2=True, shiftPixel=False, needvel=True):
def run_generation(input_borg, a_borg, a_ic, cosmo, supersample=1, do_lpt2=True, shiftPixel=False, psi_instead=False, needvel=True):
""" Generate particles and velocities from a BORG snapshot. Returns a tuple of
(positions,velocities,N,BoxSize,scale_factor)."""
@ -74,7 +74,7 @@ def run_generation(input_borg, a_borg, a_ic, cosmo, supersample=1, do_lpt2=True,
# Compute LPT scaling coefficient
D1 = cgrowth.D(a_ic)
D1_0 = D1/cgrowth.D(a_borg)
velmul = cgrowth.compute_velmul(a_ic)
velmul = cgrowth.compute_velmul(a_ic) if not psi_instead else 1
D2 = -3./7 * D1_0**2
@ -119,7 +119,7 @@ def whitify(density, L, cosmo, supergenerate=1, func='HU_WIGGLES'):
Pk = build_Pk()
Pk[0,0,0]=1
cube = CubeFT(N, L)
cube = CubeFT(L, N)
cube.density = density
density_hat = cube.rfft()
density_hat /= np.sqrt(Pk)
@ -170,15 +170,15 @@ def write_icfiles(*generated_ic, **kwargs):
supergenerate=kwargs['supergenerate']
posx,vel,density,N,L,a_ic,cosmo = generated_ic
ct.simpleWriteGadget("borg.gad", posx, velocities=vel, boxsize=L, Hubble=cosmo['h'], Omega_M=cosmo['omega_M_0'], time=a_ic)
ct.simpleWriteGadget("Data/borg.gad", posx, velocities=vel, boxsize=L, Hubble=cosmo['h'], Omega_M=cosmo['omega_M_0'], time=a_ic)
for i,c in enumerate(["x","y","z"]):
ct.writeGrafic("ic_velc%s" % c, vel[i].reshape((N,N,N)), L, a_ic, **cosmo)
ct.writeGrafic("Data/ic_velc%s" % c, vel[i].reshape((N,N,N)), L, a_ic, **cosmo)
ct.writeGrafic("ic_deltab", density, L, a_ic, **cosmo)
ct.writeGrafic("Data/ic_deltab", density, L, a_ic, **cosmo)
ct.writeWhitePhase("white.dat", whitify(density, L, cosmo, supergenerate=supergenerate))
ct.writeWhitePhase("Data/white.dat", whitify(density, L, cosmo, supergenerate=supergenerate))
with file("white_params", mode="w") as f:
with file("Data/white_params", mode="w") as f:
f.write("4\n%lg, %lg, %lg\n" % (cosmo['omega_M_0'], cosmo['omega_lambda_0'], 100*cosmo['h']))
f.write("%lg\n%lg\n-%lg\n0,0\n" % (cosmo['omega_B_0'],cosmo['ns'],cosmo['SIGMA8']))
f.write("-%lg\n1\n0\n\n\n\n\n" % L)

View File

@ -1,3 +1,4 @@
import numexpr as ne
import multiprocessing
import pyfftw
import weakref
@ -13,14 +14,14 @@ class CubeFT(object):
self.max_cpu = multiprocessing.cpu_count() if max_cpu < 0 else max_cpu
self._dhat = pyfftw.n_byte_align_empty((self.N,self.N,self.N/2+1), self.align, dtype='complex64')
self._density = pyfftw.n_byte_align_empty((self.N,self.N,self.N), self.align, dtype='float32')
self.irfft = pyfftw.FFTW(self._dhat, self._density, axes=(0,1,2), direction='FFTW_BACKWARD', threads=self.max_cpu, normalize_idft=False)
self.rfft = pyfftw.FFTW(self._density, self._dhat, axes=(0,1,2), threads=self.max_cpu, normalize_idft=False)
self._irfft = pyfftw.FFTW(self._dhat, self._density, axes=(0,1,2), direction='FFTW_BACKWARD', threads=self.max_cpu, normalize_idft=False)
self._rfft = pyfftw.FFTW(self._density, self._dhat, axes=(0,1,2), threads=self.max_cpu, normalize_idft=False)
def rfft(self):
return self.rfft()*(self.L/self.N)**3
return ne.evaluate('c*a', local_dict={'c':self._rfft(normalise_idft=False),'a':(self.L/self.N)**3})
def irfft(self):
return self.irfft()/self.L**3
return ne.evaluate('c*a', local_dict={'c':self._irfft(normalise_idft=False),'a':(1/self.L)**3})
def get_dhat(self):
return self._dhat
@ -152,16 +153,18 @@ class LagrangianPerturbation(object):
k2 = self._get_k2()
k2[0,0,0] = 1
potgen0 = lambda i: ne.evaluate('kdir**2*d/k2',local_dict={'kdir':self._kdir(i),'d':self.dhat,'k2':k2} )
potgen = lambda i,j: ne.evaluate('kdir0*kdir1*d/k2',local_dict={'kdir0':self._kdir(i),'kdir1':self._kdir(j),'d':self.dhat,'k2':k2} )
if 'lpt2_potential' not in self.cache:
print("Rebuilding potential...")
div_phi2 = np.zeros((self.N,self.N,self.N), dtype=np.float64)
for j in xrange(3):
q = self._do_irfft( self._kdir(j)**2*self.dhat / k2 ).copy()
q = self._do_irfft( potgen0(j) ).copy()
for i in xrange(j+1, 3):
div_phi2 += q * self._do_irfft( self._kdir(i)**2*self.dhat / k2 )
div_phi2 -= (self._do_irfft(self._kdir(j)*self._kdir(i)*self.dhat / k2 ) )**2
div_phi2 += q * self._do_irfft( potgen0(i) )
div_phi2 -= self._do_irfft(potgen(i,j))**2
div_phi2 *= 1/self.L**6
phi2_hat = -self._do_rfft(div_phi2) / k2
#self.cache['lpt2_potential'] = phi2_hat
del div_phi2

View File

@ -9,10 +9,10 @@ cosmo['omega_B_0']=0.049
cosmo['SIGMA8']=0.8344
cosmo['ns']=0.9624
supergen=8
supergen=2
zstart=50
astart=1/(1.+zstart)
halfPixelShift=False
if __name__=="__main__":
bic.write_icfiles(*bic.run_generation("initial_condition_borg.dat", 0.001, astart, cosmo, supersample=1, shiftPixel=halfPixelShift, do_lpt2=False), supergenerate=supergen)
bic.write_icfiles(*bic.run_generation("initial_condition_borg.dat", 0.001, astart, cosmo, supersample=2, shiftPixel=halfPixelShift, do_lpt2=False), supergenerate=supergen)

View File

@ -12,8 +12,8 @@ cosmo['SIGMA8']=0.8344
cosmo['ns']=0.9624
N0=256
doSimulation=True
simShift=True
doSimulation=False
simShift=False
snap_id=int(sys.argv[1])
astart=1/100.
@ -37,7 +37,7 @@ if doSimulation:
dsim_hat = np.fft.rfftn(dsim)*(L/N0)**3
Psim, bsim = bic.bin_power(np.abs(dsim_hat)**2/L**3, L, range=(0,1.), bins=150)
pos,_,density,N,L,_,_ = bic.run_generation("initial_density_2588.dat", 0.001, astart, cosmo, supersample=2, do_lpt2=True)
pos,_,density,N,L,_,_ = bic.run_generation("initial_density_988.dat", 0.001, astart, cosmo, supersample=1, do_lpt2=True)
dcic = ct.cicParticles(pos, L, N0)
dcic /= np.average(np.average(np.average(dcic, axis=0), axis=0), axis=0)

View File

@ -18,11 +18,15 @@ frac_electron = 1.0 # Hmmmm
frac_gas_galaxy = 0.14
mu = 1/(1-0.5*Y)
tmpp_cat={'Msun':3.29,'alpha':-0.7286211634758224,'Mstar':-23.172904033796893,'PhiStar':0.0113246633636846,'lbar':393109973.22508669}
tmpp_cat={'Msun':3.29,
'alpha':-0.7286211634758224,
'Mstar':-23.172904033796893,
'PhiStar':0.0113246633636846,
'lbar':393109973.22508669}
baryon_fraction = Omega_baryon / Omega_matter
ksz_normalization = T_cmb*sigmaT*v_unit/(lightspeed*mu*mp) * baryon_fraction
ksz_normalization = -T_cmb*sigmaT*v_unit/(lightspeed*mu*mp) * baryon_fraction
rho_mean_matter = Omega_matter * (3*(100e3/Mpc)**2/(8*np.pi*G))
Lbar = tmpp_cat['lbar'] / Mpc**3
M_over_L_galaxy = rho_mean_matter / Lbar

View File

@ -1,3 +1,4 @@
import numpy as np
from .constants import *
# -----------------------------------------------------------------------------
@ -11,27 +12,28 @@ class KSZ_Profile(object):
def __init__(self):
self.rGalaxy = 1.0
def evaluate_profile(r):
def evaluate_profile(self, r):
raise NotImplementedError("Abstract function")
def projected_profile(cos_theta,angularDistance):
def projected_profile(self, cos_theta,angularDistance):
idx = np.where(cos_theta > 0)[0]
tan_theta_2 = 1/(cos_theta[idx]**2) - 1
tan_theta_2_max = (self.rGalaxy/angularDistance)**2
tan_theta_2_min = (R_star/angularDistance)**2
tan_theta_2_min = (self.R_star/angularDistance)**2
idx0 = np.where((tan_theta_2 < tan_theta_2_max))
idx = idx[idx0]
tan_theta_2 = tan_theta_2[idx0]
tan_theta = np.sqrt(tan_theta_2)
r = (tan_theta*DA)
r = (tan_theta*angularDistance)
m,idx_mask = self.evaluate_profile(r)
idx_mask = np.append(idx_mask,np.where(tan_theta_2<tan_theta_2_min)[0])
idx_mask = np.append(idx_mask,[tan_theta_2.argmin()])
if tan_theta_2.size > 0:
idx_mask = np.append(idx_mask,[tan_theta_2.argmin()])
return idx,idx_mask,m
@ -75,7 +77,7 @@ class KSZ_Isothermal(KSZ_Profile):
self._table = x,profile
def evaluate_profile(r):
def evaluate_profile(self,r):
rho0, rGalaxy, rInner = self.rho0, self.rGalaxy, self.rInnerGalaxy
Q=rho0*2/r*np.arctan(np.sqrt((rGalaxy/r)**2 - 1))/Mpc
@ -140,7 +142,7 @@ class KSZ_NFW(KSZ_Profile):
return exp(0.971 - 0.094*log(Mvir/(1e12*MassSun)))
def evaluate_profile(r):
def evaluate_profile(self,r):
cs = self._get_concentration(self.Mvir)
rs = self.Rvir/cs