def test_subvol_gmf(Mr):
"""
"""
thr = -1.0 * np.float(Mr)
model = PrebuiltHodModelFactory("zheng07", threshold=thr, halocat="multidark", redshift=0.0)
model.new_haloprop_func_dict = {"sim_subvol": util.mk_id_column}
datsubvol = lambda x: util.mask_func(x, 0)
model.populate_mock(simname="multidark", masking_function=datsubvol, enforce_PBC=False)
# compute group richness
# These are wrong because they assume periodicity
# rich = richness(model.mock.compute_fof_group_ids())
# gmf = GMF(rich) #GMF
# print gmf
# print GMF(rich , counts = True)
galaxy_sample = model.mock.galaxy_table
x = galaxy_sample["x"]
y = galaxy_sample["y"]
z = galaxy_sample["z"]
vz = galaxy_sample["vz"]
pos = three_dim_pos_bundle(model.mock.galaxy_table, "x", "y", "z", velocity=vz, velocity_distortion_dimension="z")
b_para, b_perp = 0.2, 0.2
groups = FoFGroups(pos, b_perp, b_para, period=None, Lbox=200, num_threads="max")
gids = groups.group_ids
rich = richness(gids)
gmf = GMF(rich)
print gmf
print GMF(rich, counts=True)
return None
def main():
# Entire MultiDark Volume (Analytic xi)
cov = np.loadtxt("../data/wpxicov_dr72_bright0_mr21.0_z0.159_nj400")
print(cov.shape)
model = PrebuiltHodModelFactory('zheng07', threshold=-21)
print(model.param_dict)
model.param_dict['logM0'] = 12.59
model.param_dict['sigma_logM'] = 0.49
model.param_dict['logMmin'] = 12.78
model.param_dict['alpha'] = 1.14
model.param_dict['logM1'] = 13.99
#, 'sigma_logM': 0.39, 'logMmin': 12.79, 'alpha': 1.15, 'logM1': 13.94}
halocat = CachedHaloCatalog(simname = 'bolplanck', redshift = 0, halo_finder = 'rockstar')
model.populate_mock(halocat, enforce_PBC = True)
pos = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
tstart = time.time()
t0 = tstart
pos = pos.astype(np.float32)
x, y, z = pos[:,0] , pos[:,1] , pos[:,2]
t1 = time.time()
print("Done reading the data - time taken = {0:10.1f} seconds"
.format(t1 - t0))
print("Beginning Correlation functions calculations")
boxsize = 250
nthreads = 4
pimax = 40.0
binfile = path.join(path.dirname(path.abspath(__file__)),
"../", "bin")
autocorr = 1
numbins_to_print = 12
print("\nRunning 2-D projected correlation function wp(rp)")
results_wp = _countpairs.countpairs_wp(boxsize, pimax, nthreads,
binfile, x, y, z)
print("\n# ****** wp: first {0} bins ******* "
.format(numbins_to_print))
print("# rmin rmax rpavg wp npairs")
print("##########################################################")
for ibin in range(numbins_to_print):
items = results_wp[ibin]
print("{0:12.4f} {1:12.4f} {2:10.4f} {3:10.1f} {4:10d}"
.format(items[0], items[1], items[2], items[3], items[4]))
print("-----------------------------------------------------------")
data_wp = np.loadtxt("../data/wpxi_dr72_bright0_mr21.0_z0.159_nj400")[:,1]
print(data_wp.shape)
data_wp_error = np.sqrt(np.diag(cov)[:12])
print(data_wp_error.shape)
rbins = np.loadtxt(binfile)
rs = np.mean(rbins , axis = 1)
plt.figure(figsize=(10,10))
plt.errorbar(rs , data_wp , data_wp_error , fmt=".k" , capsize = 2)
plt.plot(rs , np.array(results_wp)[:,3])
plt.loglog()
plt.savefig("wp.pdf")
def test_GMFbinning(Mr):
""" Tests for the GMF binning scheme in order to make it sensible.
"""
gids_saved = "gids_saved.p"
if not os.path.isfile(gids_saved):
thr = -1.0 * np.float(Mr)
model = PrebuiltHodModelFactory("zheng07", threshold=thr, halocat="multidark", redshift=0.0)
model.new_haloprop_func_dict = {"sim_subvol": util.mk_id_column}
datsubvol = lambda x: util.mask_func(x, 0)
model.populate_mock(simname="multidark", masking_function=datsubvol, enforce_PBC=False)
galaxy_sample = model.mock.galaxy_table
x = galaxy_sample["x"]
y = galaxy_sample["y"]
z = galaxy_sample["z"]
vz = galaxy_sample["vz"]
pos = three_dim_pos_bundle(
model.mock.galaxy_table, "x", "y", "z", velocity=vz, velocity_distortion_dimension="z"
)
b_para, b_perp = 0.2, 0.2
groups = FoFGroups(pos, b_perp, b_para, period=None, Lbox=200, num_threads="max")
pickle.dump(groups, open(gids_saved, "wb"))
else:
groups = pickle.load(open(gids_saved, "rb"))
gids = groups.group_ids
rich = richness(gids)
# print "rich=" , rich
rbins = np.logspace(np.log10(3.0), np.log10(20), 10)
rbins = np.array([1, 2.0, 3.0, 4.0, 5.0, 6.0, 7, 9, 11, 14, 17, 20])
gmf = GMF(rich, counts=False, bins=rbins)
gmf_counts = GMF(rich, counts=True, bins=rbins)
print rbins
print gmf
print gmf_counts
fig = plt.figure(1)
sub = fig.add_subplot(111)
sub.plot(0.5 * (rbins[:-1] + rbins[1:]), gmf)
sub.set_xscale("log")
sub.set_yscale("log")
plt.show()
# print gmf
# print GMF(rich , counts = True)
return None
def test_subvol_gmf(Mr):
'''
'''
thr = -1. * np.float(Mr)
model = PrebuiltHodModelFactory('zheng07',
threshold=thr,
halocat='multidark',
redshift=0.)
model.new_haloprop_func_dict = {'sim_subvol': util.mk_id_column}
datsubvol = lambda x: util.mask_func(x, 0)
model.populate_mock(simname='multidark',
masking_function=datsubvol,
enforce_PBC=False)
#compute group richness
# These are wrong because they assume periodicity
#rich = richness(model.mock.compute_fof_group_ids())
#gmf = GMF(rich) #GMF
#print gmf
#print GMF(rich , counts = True)
galaxy_sample = model.mock.galaxy_table
x = galaxy_sample['x']
y = galaxy_sample['y']
z = galaxy_sample['z']
vz = galaxy_sample['vz']
pos = three_dim_pos_bundle(model.mock.galaxy_table,
'x',
'y',
'z',
velocity=vz,
velocity_distortion_dimension="z")
b_para, b_perp = 0.2, 0.2
groups = FoFGroups(pos,
b_perp,
b_para,
period=None,
Lbox=200,
num_threads='max')
gids = groups.group_ids
rich = richness(gids)
gmf = GMF(rich)
print gmf
print GMF(rich, counts=True)
return None
def model(theta, Mstar, prior_range = None):
'''
Given theta (HOD parameters) and stellar mass threshold Mstar,
compute wp
not sure prior_range is necessary here
'''
model = PrebuiltHodModelFactory('hearin15', threshold = Mstar,
redshift = 0.0 ,
central_assembias_strength = 0.,
satellite_assembias_strength = 0.)
model.param_dict['alphasat'] = theta[0]#1.0
model.param_dict['bcut'] = theta[1]#1.47,
model.param_dict['betacut'] = theta[2]#0.13
model.param_dict['betasat'] = theta[3]#0.859
model.param_dict['bsat'] = theta[4]#0.62
model.param_dict['mean_occupation_centrals_assembias_param1'] = theta[5] # 0
model.param_dict['mean_occupation_satellites_assembias_param1'] = theta[6] #.75
model.param_dict['scatter_model_param1'] = theta[7]#0.2
model.param_dict['smhm_beta_0'] = theta[8]# 0.43
model.param_dict['smhm_beta_a'] = theta[9]#0.18
model.param_dict['smhm_delta_0'] = theta[10]#0.56
model.param_dict['smhm_delta_a'] = theta[11]#0.18,
model.param_dict['smhm_gamma_0']= theta[12]#1.54
model.param_dict['smhm_gamma_a'] = theta[13]#2.52,
model.param_dict['smhm_m0_0'] = theta[14]# 10.72,
model.param_dict['smhm_m0_a'] = theta[15]# 0.59
model.param_dict['smhm_m1_0'] = theta[16]# 12.35,
model.param_dict['smhm_m1_a'] = theta[17]#0.3
model.populate_mock(simname = 'bolshoi', redshift = 0, halo_finder = 'rockstar')
#x = model.mock.galaxy_table['x']
#y = model.mock.galaxy_table['y']
#z = model.mock.galaxy_table['z']
#all_positions = return_xyz_formatted_array(x, y, z)
all_positions = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
wp_all = wp(all_positions, rp_bins, pi_bins,
period=model.mock.Lbox, estimator = 'Landy-Szalay', num_threads = 1)
#wp = proj_clustering(pos , rbins , pbins , cellsize = [rmax, rmax, rmax])
print 1
return rp_bin_centers, wp_all
def build_xi_nbar_gmf(Mr=21):
'''
Build "data" xi, nbar, GMF values and write to file
'''
model = PrebuiltHodModelFactory('zheng07', threshold = -1.0*np.float(Mr))
model.populate_mock(halocat = halocat , enforce_PBC = False) # population mock realization
pos = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
# write xi
rbins = hardcoded_xi_bins()
rmax = rbins.max()
approx_cell1_size = [rmax, rmax, rmax]
approx_cell2_size = approx_cell1_size
approx_cellran_size = [rmax, rmax, rmax]
period = np.array([Lbox , Lbox , Lbox])
data_xir = tpcf(
sample1, rbins, sample2 = sample2,
randoms=randoms, period = period,
max_sample_size=int(1e4), estimator='Landy-Szalay',
approx_cell1_size=approx_cell1_size,
approx_cellran_size=approx_cellran_size,
RR_precomputed = RR,
NR_precomputed = NR1)
output_file = ''.join([util.dat_dir(), 'xir.Mr', str(Mr), '.dat'])
np.savetxt(output_file, data_xir)
# write nbar values
nbar = model.mock.number_density
output_file = ''.join([util.dat_dir(), 'nbar.Mr', str(Mr), '.dat'])
np.savetxt(output_file, [nbar])
# write GMF
rich = richness(model.mock.compute_fof_group_ids())
gmf = GMF(rich) # GMF
output_file = ''.join([util.dat_dir(), 'gmf.Mr', str(Mr), '.dat'])
np.savetxt(output_file, gmf)
return None
def build_nbar_xi_gmf_cov(Mr=21):
''' Build covariance matrix for the full nbar, xi, gmf data vector
using realisations of galaxy mocks for "data" HOD parameters in the
halos from the multidark simulation. Covariance matrices for different sets of observables
can be extracted from the full covariance matrix by slicing through
the indices.
'''
nbars = []
xir = []
gmfs = []
thr = -1. * np.float(Mr)
model = PrebuiltHodModelFactory('zheng07', threshold=thr)
halocat = CachedHaloCatalog(simname='multidark',
redshift=0,
halo_finder='rockstar')
#some settings for tpcf calculations
rbins = hardcoded_xi_bins()
for i in xrange(1, 125):
print 'mock#', i
# populate the mock subvolume
model.populate_mock(halocat)
# returning the positions of galaxies
pos = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
# calculate nbar
nbars.append(len(pos) / 1000**3.)
# translate the positions of randoms to the new subbox
#calculate xi(r)
xi = tpcf(pos,
rbins,
period=model.mock.Lbox,
max_sample_size=int(2e5),
estimator='Landy-Szalay')
xir.append(xi)
# calculate gmf
nbar = len(pos) / 1000**3.
b_normal = 0.75
b = b_normal * (nbar)**(-1. / 3)
groups = pyfof.friends_of_friends(pos, b)
w = np.array([len(x) for x in groups])
gbins = gmf_bins()
gmf = np.histogram(w, gbins)[0] / (1000.**3.)
gmfs.append(gmf) # GMF
# save nbar variance
nbar_var = np.var(nbars, axis=0, ddof=1)
nbar_file = ''.join(
[util.multidat_dir(), 'abc_nbar_var.Mr',
str(Mr), '.dat'])
np.savetxt(nbar_file, [nbar_var])
# write full covariance matrix of various combinations of the data
# and invert for the likelihood evaluations
# --- covariance for all three ---
fulldatarr = np.hstack(
(np.array(nbars).reshape(len(nbars),
1), np.array(xir), np.array(gmfs)))
fullcov = np.cov(fulldatarr.T)
fullcorr = np.corrcoef(fulldatarr.T)
# and save the covariance matrix
nopoisson_file = ''.join([
util.multidat_dir(), 'abc_nbar_xi_gmf_cov.no_poisson.Mr',
str(Mr), '.dat'
])
np.savetxt(nopoisson_file, fullcov)
# and a correlation matrix
full_corr_file = ''.join(
[util.multidat_dir(), 'abc_nbar_xi_gmf_corr.Mr',
str(Mr), '.dat'])
np.savetxt(full_corr_file, fullcorr)
return None
np.log10(20.), 15)])
rmax = xi_bin.max()
# Get those randoms
num_randoms = 50 * 800000
xran = np.random.uniform(0, 1000, num_randoms)
yran = np.random.uniform(0, 1000, num_randoms)
zran = np.random.uniform(0, 1000, num_randoms)
full_randoms = np.vstack((xran, yran, zran)).T
# Get the full box mock
model = PrebuiltHodModelFactory('zheng07', threshold=-21)
halocat = CachedHaloCatalog(simname='multidark', redshift=0,
halo_finder='rockstar')
model.populate_mock(halocat, enforce_PBC=False)
pos = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
# Get full tpcf
print "getting full vol tpcf..."
xi_full_pc = tpcf(pos, xi_bin,
randoms=full_randoms,
do_auto=True, do_cross=False,
max_sample_size=int(pos.shape[0]),
estimator='Natural',
approx_cell1_size=[rmax, rmax, rmax],
approx_cellran_size=[rmax, rmax, rmax])
print "done"
Nsub = 8
# Now set up for subvol boxes
def main():
cov = np.loadtxt("../../data/wpxicov_dr72_bright0_mr21.0_z0.159_nj400")
f_MD = (1. + 71.74*10**6. / (1000.)**3.)
f_bol = (1. + 71.74*10.**6. / (250.)**3.)
print("covariance correction factor=" , f_bol/f_MD)
cov = cov*f_bol/f_MD
model.param_dict['logM0'] = 12.59
model.param_dict['sigma_logM'] = 0.49
model.param_dict['logMmin'] = 12.78
model.param_dict['alpha'] = 1.14
model.param_dict['logM1'] = 13.99
model.param_dict['mean_occupation_satellites_assembias_param1'] = 0.0
halocat = CachedHaloCatalog(simname = 'bolplanck', redshift = 0, halo_finder = 'rockstar')
model.populate_mock(halocat, enforce_PBC = True)
pos = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
print("modelnumber density=" , len(pos)/halocat.Lbox**3.)
print("data number density=" , 1.16*10**-3.)
x = model.mock.galaxy_table['x']
y = model.mock.galaxy_table['y']
z = model.mock.galaxy_table['z']
vz = model.mock.galaxy_table['vz']
# applying RSD
pos = return_xyz_formatted_array(x, y, z, velocity = vz, velocity_distortion_dimension = 'z')
# enfprcing PBC
pos = enforce_periodicity_of_box(pos, halocat.Lbox)
tstart = time.time()
t0 = tstart
pos = pos.astype(np.float32)
x, y, z = pos[:,0] , pos[:,1] , pos[:,2]
t1 = time.time()
print("Done reading the data - time taken = {0:10.1f} seconds"
.format(t1 - t0))
print("Beginning Correlation functions calculations")
boxsize = halocat.Lbox
nthreads = 4
pimax = 40.0
binfile = path.join(path.dirname(path.abspath(__file__)),
"../../", "bin")
autocorr = 1
numbins_to_print = 12
print("\nRunning 2-D projected correlation function wp(rp)")
results_wp = _countpairs.countpairs_wp(boxsize, pimax, nthreads,
binfile, x, y, z)
print("\n# ****** wp: first {0} bins ******* "
.format(numbins_to_print))
print("# rmin rmax rpavg wp npairs")
print("##########################################################")
for ibin in range(numbins_to_print):
items = results_wp[ibin]
print("{0:12.4f} {1:12.4f} {2:10.4f} {3:10.1f} {4:10d}"
.format(items[0], items[1], items[2], items[3], items[4]))
print("-----------------------------------------------------------")
data_wp = np.loadtxt("../../data/wpxi_dr72_bright0_mr21.0_z0.159_nj400")[:,1]
print(data_wp.shape)
data_wp_error = np.sqrt(np.diag(cov)[:12])
print(data_wp_error.shape)
rbins = np.loadtxt(binfile)
rs = np.mean(rbins , axis = 1)
plt.figure(figsize=(10,10))
plt.errorbar(rs , data_wp , data_wp_error , fmt=".k" , capsize = 2)
plt.plot(rs , np.array(results_wp)[:,3])
plt.loglog()
plt.savefig("wp.pdf")
def Subvolume_FullvolumeCut(N_sub, ratio=False):
''' Test the 2PCF estimates from MultiDark subvolume versus the
2PCF for the entire MultiDark volume WITHOUT periodic boundary conditions
and actual pair counts, CUT into subvolumes of the same size *AFTER*
populate mock
Parameters
----------
N_sub : (int)
Number of subvolumes to sample
'''
prettyplot()
pretty_colors = prettycolors()
pickle_file = ''.join([
'/export/bbq2/hahn/ccppabc/dump/', 'xi_subvolume_fullvolume_cut_test',
'.Nsub',
str(N_sub), '.p'
])
fig = plt.figure(1)
sub = fig.add_subplot(111)
xi_bin = xi_binedges()
# Entire MultiDark Volume (No Periodic Boundary Conditions)
model = PrebuiltHodModelFactory('zheng07', threshold=-21)
halocat = CachedHaloCatalog(simname='multidark',
redshift=0,
halo_finder='rockstar')
sub_RR = data_RR(box='md_sub')
sub_randoms = data_random(box='md_sub')
sub_NR = len(sub_randoms)
rmax = xi_bin.max()
full_approx_cell1_size = [rmax, rmax, rmax]
full_approx_cellran_size = [rmax, rmax, rmax]
model.populate_mock(halocat, enforce_PBC=False)
subvol_id = util.mk_id_column(table=model.mock.galaxy_table)
full_pos = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
# Full Volume
if os.path.isfile(pickle_file):
data_dump = pickle.load(open(pickle_file, 'rb'))
full_xi = data_dump['full_xi']
else:
model = PrebuiltHodModelFactory('zheng07', threshold=-21)
halocat = CachedHaloCatalog(simname='multidark',
redshift=0,
halo_finder='rockstar')
full_randoms = data_random(box='md_all')
full_RR = data_RR(box='md_all')
full_NR = len(full_randoms)
rmax = xi_bin.max()
full_approx_cell1_size = [rmax, rmax, rmax]
full_approx_cellran_size = [rmax, rmax, rmax]
model.populate_mock(halocat, enforce_PBC=False)
full_pos = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
full_xi = tpcf(full_pos,
xi_bin,
randoms=full_randoms,
period=None,
do_auto=True,
do_cross=False,
num_threads=5,
max_sample_size=int(full_pos.shape[0]),
estimator='Natural',
approx_cell1_size=full_approx_cell1_size,
approx_cellran_size=full_approx_cellran_size,
RR_precomputed=full_RR,
NR_precomputed=full_NR)
data_dump = {}
data_dump['full_xi'] = full_xi
if not ratio:
sub.plot(0.5 * (xi_bin[:-1] + xi_bin[1:]),
full_xi,
lw=2,
ls='-',
c='k',
label=r'Full Volume')
if os.path.isfile(pickle_file):
fullcut_xi_list = data_dump['fullcut_xi']['fullcut_xi_list']
fullcut_xi_avg = data_dump['fullcut_xi']['fullcut_xi_avg']
else:
data_dump['fullcut_xi'] = {}
fullcut_xi_list = []
fullcut_xi_tot = np.zeros(len(xi_bin) - 1)
for id in np.unique(subvol_id)[:N_sub]:
print 'Subvolume ', id
in_cut = np.where(subvol_id == id)
fullcut_pos = full_pos[in_cut]
fullcut_xi = tpcf(fullcut_pos,
xi_bin,
randoms=sub_randoms,
period=None,
do_auto=True,
do_cross=False,
num_threads=5,
max_sample_size=int(fullcut_pos.shape[0]),
estimator='Natural',
approx_cell1_size=full_approx_cell1_size,
approx_cellran_size=full_approx_cellran_size,
RR_precomputed=sub_RR,
NR_precomputed=sub_NR)
fullcut_xi_list.append(fullcut_xi)
fullcut_xi_tot += fullcut_xi
fullcut_xi_avg = fullcut_xi_tot / np.float(N_sub)
data_dump['fullcut_xi']['fullcut_xi_list'] = fullcut_xi_list
data_dump['fullcut_xi']['fullcut_xi_avg'] = fullcut_xi_avg
if not ratio:
sub.plot(0.5 * (xi_bin[:-1] + xi_bin[1:]),
fullcut_xi_avg,
lw=2,
ls='-',
c='k',
label=r'Full Volume Cut Average')
else:
sub.plot(0.5 * (xi_bin[:-1] + xi_bin[1:]),
fullcut_xi_avg / full_xi,
lw=2,
ls='-',
c='k',
label=r'Full Volume Cut Average')
if not os.path.isfile(pickle_file):
# MultiDark SubVolume (precomputed RR pairs)
sub_model = PrebuiltHodModelFactory('zheng07', threshold=-21)
sub_model.new_haloprop_func_dict = {'sim_subvol': util.mk_id_column}
sub_halocat = CachedHaloCatalog(simname='multidark',
redshift=0,
halo_finder='rockstar')
sub_RR = data_RR(box='md_sub')
sub_randoms = data_random(box='md_sub')
sub_NR = len(sub_randoms)
sub_xis_list = []
sub_xis = np.zeros(len(full_xi))
for ii in range(1, N_sub):
print 'Subvolume ', ii
# randomly sample one of the subvolumes
rint = ii #np.random.randint(1, 125)
simsubvol = lambda x: util.mask_func(x, rint)
sub_model.populate_mock(sub_halocat,
masking_function=simsubvol,
enforce_PBC=False)
pos = three_dim_pos_bundle(sub_model.mock.galaxy_table, 'x', 'y',
'z')
xi, yi, zi = util.random_shifter(rint)
temp_randoms = sub_randoms.copy()
temp_randoms[:, 0] += xi
temp_randoms[:, 1] += yi
temp_randoms[:, 2] += zi
rmax = xi_bin.max()
sub_approx_cell1_size = [rmax, rmax, rmax]
sub_approx_cellran_size = [rmax, rmax, rmax]
sub_xi = tpcf(pos,
xi_bin,
randoms=temp_randoms,
period=None,
do_auto=True,
do_cross=False,
num_threads=5,
max_sample_size=int(pos.shape[0]),
estimator='Natural',
approx_cell1_size=sub_approx_cell1_size,
approx_cellran_size=sub_approx_cellran_size,
RR_precomputed=sub_RR,
NR_precomputed=sub_NR)
label = None
if ii == N_sub - 1:
label = 'Subvolumes'
sub_xis += sub_xi
sub_xis_list.append(sub_xi)
sub_xi_avg = sub_xis / np.float(N_sub)
data_dump['Natural'] = {}
data_dump['Natural']['sub_xi_avg'] = sub_xi_avg
data_dump['Natural']['sub_xis_list'] = sub_xis_list
else:
sub_xis_list = data_dump['Natural']['sub_xis_list']
sub_xi_avg = data_dump['Natural']['sub_xi_avg']
if not os.path.isfile(pickle_file):
pickle.dump(data_dump, open(pickle_file, 'wb'))
if not ratio:
sub.plot(0.5 * (xi_bin[:-1] + xi_bin[1:]),
sub_xi_avg,
lw=2,
ls='--',
c=pretty_colors[3],
label='Subvolume')
else:
sub.plot(0.5 * (xi_bin[:-1] + xi_bin[1:]),
sub_xi_avg / full_xi,
lw=2,
ls='--',
c=pretty_colors[3],
label='Subvolume')
sub.set_xlim([0.1, 50.])
sub.set_xlabel('r', fontsize=30)
sub.set_xscale('log')
if not ratio:
sub.set_ylabel(r"$\xi \mathtt{(r)}$", fontsize=25)
sub.set_yscale('log')
else:
sub.set_ylabel(r"$\overline{\xi^\mathtt{sub}}/\xi^\mathtt{all}$",
fontsize=25)
sub.legend(loc='lower left')
if ratio:
fig_file = ''.join([
util.fig_dir(), 'test_xi_subvolume_fullvolume_cut.Nsub',
str(N_sub), '.ratio.png'
])
else:
fig_file = ''.join([
util.fig_dir(), 'test_xi_subvolume_fullvolume_cut.Nsub',
str(N_sub), '.png'
])
fig.savefig(fig_file, bbox_inches='tight', dpi=100)
plt.close()
return None
def test_precomputed_rr(Nr, Mr = 21):
'''
Mr = Luminositty threshold
Nr = Number of randoms
'''
rbins = np.logspace(-1, 1.25, 15)
rmax = rbins.max()
rbin_centers = (rbins[1:] + rbins[0:-1])/2.
halocat = CachedHaloCatalog(simname = 'bolshoi', redshift = 0)
model = PrebuiltHodModelFactory("zheng07")
model.populate_mock(halocat = halocat, enforce_PBC = False)
data = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
print data.shape
L = halocat.Lbox
xmin , ymin , zmin = 0., 0., 0.
xmax , ymax , zmax = L, L, L
num_randoms = Nr
xran = np.random.uniform(xmin, xmax, num_randoms)
yran = np.random.uniform(ymin, ymax, num_randoms)
zran = np.random.uniform(zmin, zmax, num_randoms)
randoms = np.vstack((xran, yran, zran)).T
verbose = False
num_threads = cpu_count()
period = None
approx_cell1_size = [rmax, rmax, rmax]
approx_cell2_size = approx_cell1_size
approx_cellran_size = [rmax, rmax, rmax]
normal_result = tpcf(
data, rbins, data,
randoms=randoms, period = period,
max_sample_size=int(1e4), estimator='Landy-Szalay',
approx_cell1_size=approx_cell1_size,
approx_cellran_size=approx_cellran_size)
#count data pairs
DD = npairs(
data, data, rbins, period,
verbose, num_threads,
approx_cell1_size, approx_cell2_size)
DD = np.diff(DD)
#count random pairs
RR = npairs(
randoms, randoms, rbins, period,
verbose, num_threads,
approx_cellran_size, approx_cellran_size)
RR = np.diff(RR)
#count data random pairs
DR = npairs(
data, randoms, rbins, period,
verbose, num_threads,
approx_cell1_size, approx_cell2_size)
DR = np.diff(DR)
print "DD=", DD
print "DR=", DR
print "RR=", RR
ND = len(data)
NR = len(randoms)
factor1 = ND*ND/(NR*NR)
factor2 = ND*NR/(NR*NR)
mult = lambda x,y: x*y
xi_LS = mult(1.0/factor1,DD/RR) - mult(1.0/factor2,2.0*DR/RR) + 1.0
print "xi=" , xi_LS
print "normal=" , normal_result
result_with_RR_precomputed = tpcf(
data, rbins, data,
randoms=randoms, period = period,
max_sample_size=int(1e5), estimator='Landy-Szalay',
approx_cell1_size=approx_cell1_size,
approx_cellran_size=approx_cellran_size,
RR_precomputed = RR,
NR_precomputed = NR)
print "xi_pre=" , result_with_RR_precomputed
from halotools.empirical_models import PrebuiltHodModelFactory
# The "fiducial" stellar mass threshold is 10**10.5
model = PrebuiltHodModelFactory('hearin15', threshold = 10.5, redshift = 0.,
central_assembias_strength = 1,
satellite_assembias_strength = 1)
model.param_dict['scatter_model_param1'] = 0.4 # This is the "fiducial" scatter used throughout the paper
baseline_model = PrebuiltHodModelFactory('leauthaud11', threshold = 10.5, redshift = 0.)
baseline_model.param_dict['scatter_model_param1'] = 0.4
#base_model
baseline_model.populate_mock()
pos = three_dim_pos_bundle(baseline_model.mock.galaxy_table, 'x', 'y', 'z')
wp_base = proj_clustering(pos , rbins , pbins , cellsize = [rmax, rmax, rmax])
#central assembly bias only
model.param_dict['mean_occupation_satellites_assembias_param1'] = 0
model.param_dict['mean_occupation_centrals_assembias_param1'] = 1
model.populate_mock()
pos = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
wp_cen_only = proj_clustering(pos , rbins , pbins , cellsize = [rmax, rmax, rmax])
# Satellite assembias only
model.param_dict['mean_occupation_satellites_assembias_param1'] = 1
model.param_dict['mean_occupation_centrals_assembias_param1'] = 0
def main():
###############################Multi-Dark###########################
model = PrebuiltHodModelFactory('zheng07', threshold=-21)
halocat = CachedHaloCatalog(simname = 'multidark', redshift = 0, halo_finder = 'rockstar')
t0 = time.time()
model.populate_mock(halocat, enforce_PBC = False)
print(time.time() - t0)
t0 = time.time()
model.populate_mock(halocat, enforce_PBC = False)
print(time.time() - t0)
pos = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
nthreads = 1
binfile = path.join(path.dirname(path.abspath(__file__)),
"/home/mj/Corrfunc/xi_theory/tests/", "bins")
autocorr = 1
pos = pos.astype(np.float32)
x , y , z = pos[:,0] , pos[:,1] , pos[:,2]
DD = _countpairs.countpairs(autocorr, nthreads, binfile, x, y, z,
x , y , z)
ND = len(pos)
NR = 50*800000
DD = np.array(DD)[:,3]
num_randoms = 50 * 800000
xran = np.random.uniform(0, 1000, num_randoms)
yran = np.random.uniform(0, 1000, num_randoms)
zran = np.random.uniform(0, 1000, num_randoms)
randoms = np.vstack((xran, yran, zran)).T
randoms = randoms.astype(np.float32)
xran = randoms[:,0]
yran = randoms[:,1]
zran = randoms[:,2]
results_RR = _countpairs.countpairs(autocorr, nthreads, binfile, xran, yran, zran,
xran, yran, zran)
RR = np.array(results_RR)[:,3]
factor1 = 1.*ND*ND/(NR*NR)
mult = lambda x,y: x*y
xi_MD = mult(1.0/factor1 , DD/RR) - 1.
print(xi_MD)
###############################Subvolume of Multi-Dark###########################
num_srandoms = 50 * 8000
xran = np.random.uniform(0, 200, num_srandoms)
yran = np.random.uniform(0, 200, num_srandoms)
zran = np.random.uniform(0, 200, num_srandoms)
randoms = np.vstack((xran, yran, zran)).T
randoms = randoms.astype(np.float32)
xran = randoms[:,0]
yran = randoms[:,1]
zran = randoms[:,2]
results_RR = _countpairs.countpairs(autocorr, nthreads, binfile, xran, yran, zran,
xran, yran, zran)
RR_sub = np.array(results_RR)[:,3]
import util
sub_model = PrebuiltHodModelFactory('zheng07' , threshold = -21)
sub_model.new_haloprop_func_dict = {'sim_subvol': util.mk_id_column}
sub_halocat = CachedHaloCatalog(simname='multidark', redshift=0, halo_finder='rockstar')
xi_subs = []
for i in range(10):
simsubvol = lambda x: util.mask_func(x, i)
sub_model.populate_mock(sub_halocat, masking_function=simsubvol, enforce_PBC=False)
sub_pos = three_dim_pos_bundle(sub_model.mock.galaxy_table, 'x', 'y', 'z')
nthreads = 1
binfile = path.join(path.dirname(path.abspath(__file__)),
"/home/mj/Corrfunc/xi_theory/tests/", "bins")
autocorr = 1
sub_pos = sub_pos.astype(np.float32)
x , y , z = sub_pos[:,0] , sub_pos[:,1] , sub_pos[:,2]
DD_sub = _countpairs.countpairs(autocorr, nthreads, binfile, x, y, z,
x, y, z)
ND_sub = len(sub_pos)
NR_sub = 50 * 8000
DD_sub = np.array(DD_sub)[:,3]
factor1 = 1.*ND_sub*ND_sub/(NR_sub*NR_sub)
mult = lambda x,y: x*y
xi_n = mult(1.0/factor1 , DD_sub/RR_sub) - 1.
xi_subs.append(xi_n)
xi_subs = np.array(xi_subs)
np.savetxt("xi_subs.dat" , xi_subs)
import matplotlib.pyplot as plt
from ChangTools.plotting import prettyplot
from ChangTools.plotting import prettycolors
binfile = path.join(path.dirname(path.abspath(__file__)),
"/home/mj/Corrfunc/xi_theory/tests/", "bins")
rbins = np.loadtxt(binfile)
rbins_centers = np.mean(rbins , axis = 1)
xi_subs = np.loadtxt("xi_subs.dat")
fig = plt.figure(figsize=(10,10))
ax = fig.add_subplot(111)
for i in range(10):
ax.semilogx(rbins_centers , xi_subs[i,:] / xi_MD , alpha = 0.2)
plt.xlabel(r"$r$" , fontsize = 20)
plt.ylabel(r"$\xi_{\rm subvolume}(r) / \xi_{\rm MD}(r)$" , fontsize = 20)
plt.savefig("xi_ratios.pdf")
def Subvolume_FullvolumeCut(N_sub, ratio=False):
''' Test the 2PCF estimates from MultiDark subvolume versus the
2PCF for the entire MultiDark volume WITHOUT periodic boundary conditions
and actual pair counts, CUT into subvolumes of the same size *AFTER*
populate mock
Parameters
----------
N_sub : (int)
Number of subvolumes to sample
'''
prettyplot()
pretty_colors = prettycolors()
pickle_file = ''.join([
'/export/bbq2/hahn/ccppabc/dump/',
'xi_subvolume_fullvolume_cut_test',
'.Nsub', str(N_sub),
'.p'])
fig = plt.figure(1)
sub = fig.add_subplot(111)
xi_bin = xi_binedges()
# Entire MultiDark Volume (No Periodic Boundary Conditions)
model = PrebuiltHodModelFactory('zheng07', threshold=-21)
halocat = CachedHaloCatalog(simname = 'multidark', redshift = 0, halo_finder = 'rockstar')
sub_RR = data_RR(box='md_sub')
sub_randoms = data_random(box='md_sub')
sub_NR = len(sub_randoms)
rmax = xi_bin.max()
full_approx_cell1_size = [rmax , rmax , rmax]
full_approx_cellran_size = [rmax , rmax , rmax]
model.populate_mock(halocat, enforce_PBC=False)
subvol_id = util.mk_id_column(table=model.mock.galaxy_table)
full_pos = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
# Full Volume
if os.path.isfile(pickle_file):
data_dump = pickle.load(open(pickle_file, 'rb'))
full_xi = data_dump['full_xi']
else:
model = PrebuiltHodModelFactory('zheng07', threshold=-21)
halocat = CachedHaloCatalog(simname = 'multidark', redshift = 0, halo_finder = 'rockstar')
full_randoms = data_random(box='md_all')
full_RR = data_RR(box='md_all')
full_NR = len(full_randoms)
rmax = xi_bin.max()
full_approx_cell1_size = [rmax , rmax , rmax]
full_approx_cellran_size = [rmax , rmax , rmax]
model.populate_mock(halocat, enforce_PBC=False)
full_pos = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
full_xi = tpcf(
full_pos, xi_bin,
randoms=full_randoms, period=None,
do_auto=True,
do_cross=False,
num_threads=5,
max_sample_size=int(full_pos.shape[0]), estimator='Natural',
approx_cell1_size=full_approx_cell1_size,
approx_cellran_size=full_approx_cellran_size,
RR_precomputed = full_RR,
NR_precomputed = full_NR)
data_dump = {}
data_dump['full_xi'] = full_xi
if not ratio:
sub.plot(0.5*(xi_bin[:-1]+xi_bin[1:]), full_xi,
lw=2, ls='-', c='k', label=r'Full Volume')
if os.path.isfile(pickle_file):
fullcut_xi_list = data_dump['fullcut_xi']['fullcut_xi_list']
fullcut_xi_avg = data_dump['fullcut_xi']['fullcut_xi_avg']
else:
data_dump['fullcut_xi'] = {}
fullcut_xi_list = []
fullcut_xi_tot = np.zeros(len(xi_bin)-1)
for id in np.unique(subvol_id)[:N_sub]:
print 'Subvolume ', id
in_cut = np.where(subvol_id == id)
fullcut_pos = full_pos[in_cut]
fullcut_xi = tpcf(
fullcut_pos, xi_bin,
randoms=sub_randoms, period=None,
do_auto=True,
do_cross=False,
num_threads=5,
max_sample_size=int(fullcut_pos.shape[0]), estimator='Natural',
approx_cell1_size=full_approx_cell1_size,
approx_cellran_size=full_approx_cellran_size,
RR_precomputed=sub_RR,
NR_precomputed=sub_NR)
fullcut_xi_list.append(fullcut_xi)
fullcut_xi_tot += fullcut_xi
fullcut_xi_avg = fullcut_xi_tot / np.float(N_sub)
data_dump['fullcut_xi']['fullcut_xi_list']= fullcut_xi_list
data_dump['fullcut_xi']['fullcut_xi_avg']= fullcut_xi_avg
if not ratio:
sub.plot(0.5*(xi_bin[:-1]+xi_bin[1:]), fullcut_xi_avg,
lw=2, ls='-', c='k', label=r'Full Volume Cut Average')
else:
sub.plot(0.5*(xi_bin[:-1]+xi_bin[1:]), fullcut_xi_avg/full_xi,
lw=2, ls='-', c='k', label=r'Full Volume Cut Average')
if not os.path.isfile(pickle_file):
# MultiDark SubVolume (precomputed RR pairs)
sub_model = PrebuiltHodModelFactory('zheng07', threshold=-21)
sub_model.new_haloprop_func_dict = {'sim_subvol': util.mk_id_column}
sub_halocat = CachedHaloCatalog(simname = 'multidark', redshift = 0, halo_finder = 'rockstar')
sub_RR = data_RR(box='md_sub')
sub_randoms = data_random(box='md_sub')
sub_NR = len(sub_randoms)
sub_xis_list = []
sub_xis = np.zeros(len(full_xi))
for ii in range(1,N_sub):
print 'Subvolume ', ii
# randomly sample one of the subvolumes
rint = ii #np.random.randint(1, 125)
simsubvol = lambda x: util.mask_func(x, rint)
sub_model.populate_mock(sub_halocat, masking_function=simsubvol, enforce_PBC=False)
pos = three_dim_pos_bundle(sub_model.mock.galaxy_table, 'x', 'y', 'z')
xi, yi , zi = util.random_shifter(rint)
temp_randoms = sub_randoms.copy()
temp_randoms[:,0] += xi
temp_randoms[:,1] += yi
temp_randoms[:,2] += zi
rmax = xi_bin.max()
sub_approx_cell1_size = [rmax , rmax , rmax]
sub_approx_cellran_size = [rmax , rmax , rmax]
sub_xi = tpcf(
pos, xi_bin,
randoms=temp_randoms,
period = None,
do_auto=True,
do_cross=False,
num_threads=5,
max_sample_size=int(pos.shape[0]),
estimator='Natural',
approx_cell1_size = sub_approx_cell1_size,
approx_cellran_size = sub_approx_cellran_size,
RR_precomputed=sub_RR,
NR_precomputed=sub_NR)
label = None
if ii == N_sub - 1:
label = 'Subvolumes'
sub_xis += sub_xi
sub_xis_list.append(sub_xi)
sub_xi_avg = sub_xis/np.float(N_sub)
data_dump['Natural'] = {}
data_dump['Natural']['sub_xi_avg'] = sub_xi_avg
data_dump['Natural']['sub_xis_list'] = sub_xis_list
else:
sub_xis_list = data_dump['Natural']['sub_xis_list']
sub_xi_avg = data_dump['Natural']['sub_xi_avg']
if not os.path.isfile(pickle_file):
pickle.dump(data_dump, open(pickle_file, 'wb'))
if not ratio:
sub.plot(0.5*(xi_bin[:-1]+xi_bin[1:]), sub_xi_avg,
lw=2, ls='--', c=pretty_colors[3], label='Subvolume')
else:
sub.plot(0.5*(xi_bin[:-1]+xi_bin[1:]), sub_xi_avg/full_xi,
lw=2, ls='--', c=pretty_colors[3], label='Subvolume')
sub.set_xlim([0.1, 50.])
sub.set_xlabel('r', fontsize=30)
sub.set_xscale('log')
if not ratio:
sub.set_ylabel(r"$\xi \mathtt{(r)}$", fontsize=25)
sub.set_yscale('log')
else:
sub.set_ylabel(r"$\overline{\xi^\mathtt{sub}}/\xi^\mathtt{all}$", fontsize=25)
sub.legend(loc='lower left')
if ratio:
fig_file = ''.join([util.fig_dir(), 'test_xi_subvolume_fullvolume_cut.Nsub', str(N_sub), '.ratio.png'])
else:
fig_file = ''.join([util.fig_dir(), 'test_xi_subvolume_fullvolume_cut.Nsub', str(N_sub), '.png'])
fig.savefig(fig_file, bbox_inches='tight', dpi=100)
plt.close()
return None
def build_MCMC_cov_nbar_xi_gmf(Mr=21, b_normal=0.25):
''' Build covariance matrix used in MCMC for the full nbar, xi, gmf data vector
using realisations of galaxy mocks for "data" HOD parameters in the
halos from the other subvolumes (subvolume 1 to subvolume 125) of
the simulation. Covariance matrices for different sets of observables
can be extracted from the full covariance matrix by slicing through
the indices.
'''
nbars = []
xir = []
gmfs = []
thr = -1. * np.float(Mr)
model = PrebuiltHodModelFactory('zheng07', threshold=thr)
halocat = CachedHaloCatalog(simname='multidark',
redshift=0,
halo_finder='rockstar')
###model.new_haloprop_func_dict = {'sim_subvol': util.mk_id_column}
#some settings for tpcf calculations
rbins = xi_binedges()
rmax = rbins.max()
approx_cell1_size = [rmax, rmax, rmax]
approx_cellran_size = [rmax, rmax, rmax]
#load randoms and RRs
randoms = data_random(box='md_sub')
RR = data_RR(box='md_sub')
NR = len(randoms)
for i in xrange(1, 125):
print 'mock#', i
# populate the mock subvolume
###mocksubvol = lambda x: util.mask_func(x, i)
###model.populate_mock(halocat,
### masking_function=mocksubvol,
### enforce_PBC=False)
model.populate_mock(halocat)
# returning the positions of galaxies in the entire volume
pos = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
# masking out the galaxies outside the subvolume i
pos = util.mask_galaxy_table(pos, i)
# calculate nbar
print "shape of pos", pos.shape
nbars.append(len(pos) / 200**3.)
# translate the positions of randoms to the new subbox
xi0, yi0, zi0 = util.random_shifter(i)
temp_randoms = randoms.copy()
temp_randoms[:, 0] += xi0
temp_randoms[:, 1] += yi0
temp_randoms[:, 2] += zi0
#calculate xi(r)
xi = tpcf(pos,
rbins,
pos,
randoms=temp_randoms,
period=None,
max_sample_size=int(3e5),
estimator='Natural',
approx_cell1_size=approx_cell1_size,
approx_cellran_size=approx_cellran_size,
RR_precomputed=RR,
NR_precomputed=NR)
xir.append(xi)
# calculate gmf
nbar = len(pos) / 200**3.
b = b_normal * (nbar)**(-1. / 3)
groups = pyfof.friends_of_friends(pos, b)
w = np.array([len(x) for x in groups])
gbins = gmf_bins()
gmf = np.histogram(w, gbins)[0] / 200.**3.
gmfs.append(gmf)
# save nbar variance
nbar_var = np.var(nbars, axis=0, ddof=1)
nbar_file = ''.join([util.obvs_dir(), 'nbar_var.Mr', str(Mr), '.dat'])
np.savetxt(nbar_file, [nbar_var])
# write full covariance matrix of various combinations of the data
# and invert for the likelihood evaluations
# --- covariance for all three ---
fulldatarr = np.hstack(
(np.array(nbars).reshape(len(nbars),
1), np.array(xir), np.array(gmfs)))
fullcov = np.cov(fulldatarr.T)
fullcorr = np.corrcoef(fulldatarr.T)
# and save the covariance matrix
nopoisson_file = ''.join([
util.obvs_dir(), 'MCMC.nbar_xi_gmf_cov', '.no_poisson', '.Mr',
str(Mr), '.bnorm',
str(round(b_normal, 2)), '.dat'
])
np.savetxt(nopoisson_file, fullcov)
return None
def _sum_stat(self, theta, prior_range=None, observables=['nbar', 'gmf']):
'''
Given theta, sum_stat calculates the observables from our forward model
Parameters
----------
theta : (self explanatory)
prior_range : If specified, checks to make sure that theta is within the prior range.
'''
self.model.param_dict['logM0'] = theta[0]
self.model.param_dict['sigma_logM'] = np.exp(theta[1])
self.model.param_dict['logMmin'] = theta[2]
self.model.param_dict['alpha'] = theta[3]
self.model.param_dict['logM1'] = theta[4]
rbins = xi_binedges()
rmax = rbins.max()
approx_cell1_size = [rmax, rmax, rmax]
approx_cellran_size = [rmax, rmax, rmax]
if prior_range is None:
self.model.populate_mock(self.halocat, enforce_PBC=False)
pos = three_dim_pos_bundle(self.model.mock.galaxy_table, 'x', 'y',
'z')
obvs = []
for obv in observables:
if obv == 'nbar':
obvs.append(len(pos) /
1000.**3.) # nbar of the galaxy catalog
elif obv == 'gmf':
nbar = len(pos) / 1000**3.
b = self.b_normal * (nbar)**(-1. / 3)
groups = pyfof.friends_of_friends(pos, b)
w = np.array([len(x) for x in groups])
gbins = data_gmf_bins()
gmf = np.histogram(w, gbins)[0] / (1000.**3.)
obvs.append(gmf)
elif obv == 'xi':
greek_xi = tpcf(pos,
rbins,
pos,
randoms=randoms,
period=None,
max_sample_size=int(3e5),
estimator='Natural',
approx_cell1_size=approx_cell1_size,
approx_cellran_size=approx_cellran_size,
RR_precomputed=self.RR,
NR_precomputed=self.NR)
obvs.append(greek_xi)
else:
raise NotImplementedError(
'Only nbar 2pcf, gmf implemented so far')
return obvs
else:
if np.all((prior_range[:, 0] < theta)
& (theta < prior_range[:, 1])):
# if all theta_i is within prior range ...
try:
self.model.populate_mock(self.halocat)
pos = three_dim_pos_bundle(self.model.mock.galaxy_table,
'x', 'y', 'z')
obvs = []
for obv in observables:
if obv == 'nbar':
obvs.append(len(pos) /
1000**3.) # nbar of the galaxy catalog
elif obv == 'gmf':
nbar = len(pos) / 1000**3.
b = self.b_normal * (nbar)**(-1. / 3)
groups = pyfof.friends_of_friends(pos, b)
w = np.array([len(x) for x in groups])
gbins = data_gmf_bins()
gmf = np.histogram(w, gbins)[0] / (1000.**3.)
obvs.append(gmf)
elif obv == 'xi':
greek_xi = tpcf(
pos,
rbins,
pos,
randoms=randoms,
period=None,
max_sample_size=int(3e5),
estimator='Natural',
approx_cell1_size=approx_cell1_size,
approx_cellran_size=approx_cellran_size,
RR_precomputed=self.RR,
NR_precomputed=self.NR)
obvs.append(greek_xi)
else:
raise NotImplementedError(
'Only nbar, tpcf, and gmf are implemented so far'
)
return obvs
except ValueError:
obvs = []
for obv in observables:
if obv == 'nbar':
obvs.append(10.)
elif obv == 'gmf':
bins = data_gmf_bins()
obvs.append(np.ones_like(bins)[:-1] * 1000.)
elif obv == 'xi':
obvs.append(np.zeros(len(xi_binedges()[:-1])))
return obvs
else:
obvs = []
for obv in observables:
if obv == 'nbar':
obvs.append(10.)
elif obv == 'gmf':
bins = data_gmf_bins()
obvs.append(np.ones_like(bins)[:-1] * 1000.)
elif obv == 'xi':
obvs.append(np.zeros(len(xi_binedges()[:-1])))
return obvs
def _sum_stat(self, theta, prior_range=None, observables=['nbar', 'gmf']):
'''
Given theta, sum_stat calculates the observables from our forward model
Parameters
----------
theta : (self explanatory)
prior_range : If specified, checks to make sure that theta is within the prior range.
'''
self.model.param_dict['logM0'] = theta[0]
self.model.param_dict['sigma_logM'] = np.exp(theta[1])
self.model.param_dict['logMmin'] = theta[2]
self.model.param_dict['alpha'] = theta[3]
self.model.param_dict['logM1'] = theta[4]
rbins = xi_binedges()
rmax = rbins.max()
approx_cell1_size = [rmax , rmax , rmax]
approx_cellran_size = [rmax , rmax , rmax]
if prior_range is None:
self.model.populate_mock(self.halocat)
pos =three_dim_pos_bundle(self.model.mock.galaxy_table, 'x', 'y', 'z')
obvs = []
for obv in observables:
if obv == 'nbar':
obvs.append(len(pos) / 1000.**3.) # nbar of the galaxy catalog
elif obv == 'gmf':
nbar = len(pos) / 1000**3.
b = self.b_normal * (nbar)**(-1./3)
groups = pyfof.friends_of_friends(pos , b)
w = np.array([len(x) for x in groups])
gbins =data_gmf_bins()
gmf = np.histogram(w , gbins)[0] / (1000.**3.)
obvs.append(gmf)
elif obv == 'xi':
greek_xi = tpcf(
pos, rbins,
period=self.model.mock.Lbox,
max_sample_size=int(3e5), estimator='Natural',
approx_cell1_size=approx_cell1_size)
obvs.append(greek_xi)
else:
raise NotImplementedError('Only nbar 2pcf, gmf implemented so far')
return obvs
else:
if np.all((prior_range[:,0] < theta) & (theta < prior_range[:,1])):
# if all theta_i is within prior range ...
try:
self.model.populate_mock(self.halocat)
pos=three_dim_pos_bundle(self.model.mock.galaxy_table, 'x', 'y', 'z')
obvs = []
for obv in observables:
if obv == 'nbar':
obvs.append(len(pos) / 1000**3.) # nbar of the galaxy catalog
elif obv == 'gmf':
nbar = len(pos) / 1000**3.
b = self.b_normal * (nbar)**(-1./3)
groups = pyfof.friends_of_friends(pos , b)
w = np.array([len(x) for x in groups])
gbins =data_gmf_bins()
gmf = np.histogram(w , gbins)[0] / (1000.**3.)
obvs.append(gmf)
elif obv == 'xi':
greek_xi = tpcf(
pos, rbins, period=self.model.mock.Lbox,
max_sample_size=int(3e5), estimator='Natural',
approx_cell1_size=approx_cell1_size)
obvs.append(greek_xi)
else:
raise NotImplementedError('Only nbar, tpcf, and gmf are implemented so far')
return obvs
except ValueError:
obvs = []
for obv in observables:
if obv == 'nbar':
obvs.append(10.)
elif obv == 'gmf':
bins = data_gmf_bins()
obvs.append(np.ones_like(bins)[:-1]*1000.)
elif obv == 'xi':
obvs.append(np.zeros(len(xi_binedges()[:-1])))
return obvs
else:
obvs = []
for obv in observables:
if obv == 'nbar':
obvs.append(10.)
elif obv == 'gmf':
bins = data_gmf_bins()
obvs.append(np.ones_like(bins)[:-1]*1000.)
elif obv == 'xi':
obvs.append(np.zeros(len(xi_binedges()[:-1])))
return obvs
def _sum_stat(self, theta, prior_range=None, observables=['nbar', 'gmf']):
'''
Given theta, sum_stat calculates the observables from our forward model
Parameters
----------
theta : (self explanatory)
prior_range : If specified, checks to make sure that theta is within the prior range.
'''
self.model.param_dict['logM0'] = theta[0]
self.model.param_dict['sigma_logM'] = np.exp(theta[1])
self.model.param_dict['logMmin'] = theta[2]
self.model.param_dict['alpha'] = theta[3]
self.model.param_dict['logM1'] = theta[4]
rbins = xi_binedges()
rmax = rbins.max()
period = None
approx_cell1_size = [rmax , rmax , rmax]
approx_cellran_size = [rmax , rmax , rmax]
if prior_range is None:
rint = np.random.randint(1, 125)
####simsubvol = lambda x: util.mask_func(x, rint)
####self.model.populate_mock(self.halocat,
#### masking_function=simsubvol,
#### enforce_PBC=False)
self.model.populate_mock(self.halocat)
pos =three_dim_pos_bundle(self.model.mock.galaxy_table, 'x', 'y', 'z')
pos = util.mask_galaxy_table(pos , rint)
xi , yi , zi = util.random_shifter(rint)
temp_randoms = self.randoms.copy()
temp_randoms[:,0] += xi
temp_randoms[:,1] += yi
temp_randoms[:,2] += zi
obvs = []
for obv in observables:
if obv == 'nbar':
obvs.append(len(pos) / 200**3.) # nbar of the galaxy catalog
elif obv == 'gmf':
#compute group richness
nbar = len(pos) / 200**3.
b = self.b_normal * (nbar)**(-1./3)
groups = pyfof.friends_of_friends(pos , b)
w = np.array([len(x) for x in groups])
gbins = data_gmf_bins()
gmf = np.histogram(w , gbins)[0] / (200.**3.)
obvs.append(gmf)
elif obv == 'xi':
greek_xi = tpcf(
pos, rbins, pos,
randoms=temp_randoms, period = period,
max_sample_size=int(1e5), estimator='Natural',
approx_cell1_size=approx_cell1_size,
approx_cellran_size=approx_cellran_size,
RR_precomputed = self.RR,
NR_precomputed = self.NR)
obvs.append(greek_xi)
else:
raise NotImplementedError('Only nbar 2pcf, gmf implemented so far')
return obvs
else:
if np.all((prior_range[:,0] < theta) & (theta < prior_range[:,1])):
# if all theta_i is within prior range ...
try:
rint = np.random.randint(1, 125)
simsubvol = lambda x: util.mask_func(x, rint)
self.model.populate_mock(self.halocat,
masking_function=simsubvol,
enforce_PBC=False)
pos =three_dim_pos_bundle(self.model.mock.galaxy_table, 'x', 'y', 'z')
#imposing mask on the galaxy table
pos = util.mask_galaxy_table(pos , rint)
xi , yi , zi = util.random_shifter(rint)
temp_randoms = self.randoms.copy()
temp_randoms[:,0] += xi
temp_randoms[:,1] += yi
temp_randoms[:,2] += zi
obvs = []
for obv in observables:
if obv == 'nbar':
obvs.append(len(pos) / 200**3.) # nbar of the galaxy catalog
elif obv == 'gmf':
nbar = len(pos) / 200**3.
b = self.b_normal * (nbar)**(-1./3)
groups = pyfof.friends_of_friends(pos , b)
w = np.array([len(x) for x in groups])
gbins =data_gmf_bins()
gmf = np.histogram(w , gbins)[0] / (200.**3.)
obvs.append(gmf)
elif obv == 'xi':
greek_xi = tpcf(
pos, rbins, pos,
randoms=temp_randoms, period = period,
max_sample_size=int(1e5), estimator='Natural',
approx_cell1_size=approx_cell1_size,
approx_cellran_size=approx_cellran_size,
RR_precomputed = self.RR,
NR_precomputed = self.NR)
obvs.append(greek_xi)
else:
raise NotImplementedError('Only nbar, tpcf, and gmf are implemented so far')
return obvs
except ValueError:
obvs = []
for obv in observables:
if obv == 'nbar':
obvs.append(10.)
elif obv == 'gmf':
bins = data_gmf_bins()
obvs.append(np.ones_like(bins)[:-1]*1000.)
elif obv == 'xi':
obvs.append(np.zeros(len(xi_binedges()[:-1])))
return obvs
else:
obvs = []
for obv in observables:
if obv == 'nbar':
obvs.append(10.)
elif obv == 'gmf':
bins = data_gmf_bins()
obvs.append(np.ones_like(bins)[:-1]*1000.)
elif obv == 'xi':
obvs.append(np.zeros(len(xi_binedges()[:-1])))
return obvs
def test_GMFbinning(Mr):
''' Tests for the GMF binning scheme in order to make it sensible.
'''
gids_saved = 'gids_saved.p'
if not os.path.isfile(gids_saved):
thr = -1. * np.float(Mr)
model = PrebuiltHodModelFactory('zheng07',
threshold=thr,
halocat='multidark',
redshift=0.)
model.new_haloprop_func_dict = {'sim_subvol': util.mk_id_column}
datsubvol = lambda x: util.mask_func(x, 0)
model.populate_mock(simname='multidark',
masking_function=datsubvol,
enforce_PBC=False)
galaxy_sample = model.mock.galaxy_table
x = galaxy_sample['x']
y = galaxy_sample['y']
z = galaxy_sample['z']
vz = galaxy_sample['vz']
pos = three_dim_pos_bundle(model.mock.galaxy_table,
'x',
'y',
'z',
velocity=vz,
velocity_distortion_dimension="z")
b_para, b_perp = 0.2, 0.2
groups = FoFGroups(pos,
b_perp,
b_para,
period=None,
Lbox=200,
num_threads='max')
pickle.dump(groups, open(gids_saved, 'wb'))
else:
groups = pickle.load(open(gids_saved, 'rb'))
gids = groups.group_ids
rich = richness(gids)
#print "rich=" , rich
rbins = np.logspace(np.log10(3.), np.log10(20), 10)
rbins = np.array([1, 2., 3., 4., 5., 6., 7, 9, 11, 14, 17, 20])
gmf = GMF(rich, counts=False, bins=rbins)
gmf_counts = GMF(rich, counts=True, bins=rbins)
print rbins
print gmf
print gmf_counts
fig = plt.figure(1)
sub = fig.add_subplot(111)
sub.plot(0.5 * (rbins[:-1] + rbins[1:]), gmf)
sub.set_xscale('log')
sub.set_yscale('log')
plt.show()
#print gmf
#print GMF(rich , counts = True)
return None
def richness(group_id):
'''
Calculate the richness of a group given group_ids of galaxies. Uses astropy.table module
'''
gals = Table()
gals['groupid'] = group_id
gals['dummy'] = 1
grouped_table = gals.group_by('groupid')
grp_richness = grouped_table['dummy'].groups.aggregate(np.sum)
return grp_richness
model = PrebuiltHodModelFactory('zheng07', threshold=-21)
halocat = CachedHaloCatalog(simname = 'multidark', redshift = 0, halo_finder = 'rockstar')
model.populate_mock(halocat)
pos =three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
#groups = FoFGroups(pos, 0.75, 0.75, Lbox = model.mock.Lbox, num_threads=1)
#gids = groups.group_ids
#print richness(gids)
import time
a = time.time()
groups = pyfof.friends_of_friends(pos, 0.75*(len(pos)/1000**3.)**(-1./3))
w = np.array([len(x) for x in groups])
bins = np.array([2.,3.,4.,5.,6.,7,9,11,14,17,20])
def Subvolume_Analytic(N_sub, ratio=False):
''' Test the 2PCF estimates from MultiDark subvolume versus the
analytic 2PCF for the entire MultiDark volume
Parameters
----------
N_sub : (int)
Number of subvolumes to sample
'''
prettyplot()
pretty_colors = prettycolors()
pickle_file = ''.join([
'/export/bbq2/hahn/ccppabc/dump/',
'xi_subvolume_test',
'.Nsub', str(N_sub),
'.p'])
fig = plt.figure(1)
sub = fig.add_subplot(111)
xi_bin = xi_binedges()
if os.path.isfile(pickle_file):
data_dump = pickle.load(open(pickle_file, 'rb'))
full_xi = data_dump['full_xi']
else:
# Entire MultiDark Volume (Analytic xi)
model = PrebuiltHodModelFactory('zheng07', threshold=-21)
halocat = CachedHaloCatalog(simname = 'multidark', redshift = 0, halo_finder = 'rockstar')
model.populate_mock(halocat)
pos = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
# while the estimator claims to be Landy-Szalay, I highly suspect it
# actually uses Landy-Szalay since DR pairs cannot be calculated from
# analytic randoms
full_xi = tpcf(pos, xi_bin, period=model.mock.Lbox, max_sample_size=int(2e5), estimator='Landy-Szalay', num_threads=1)
data_dump = {}
data_dump['full_xi'] = full_xi
if not ratio:
sub.plot(0.5*(xi_bin[:-1]+xi_bin[1:]), full_xi, lw=2, ls='-', c='k', label=r'Analytic $\xi$ Entire Volume')
if not os.path.isfile(pickle_file):
# MultiDark SubVolume (precomputed RR pairs)
sub_model = PrebuiltHodModelFactory('zheng07', threshold=-21)
sub_model.new_haloprop_func_dict = {'sim_subvol': util.mk_id_column}
sub_halocat = CachedHaloCatalog(simname = 'multidark', redshift = 0, halo_finder = 'rockstar')
RR = data_RR()
randoms = data_random()
NR = len(randoms)
for method in ['Landy-Szalay', 'Natural']:
if method == 'Landy-Szalay':
iii = 3
elif method == 'Natural':
iii = 5
if not os.path.isfile(pickle_file):
sub_xis_list = []
sub_xis = np.zeros(len(full_xi))
for ii in range(1,N_sub+1):
# randomly sample one of the subvolumes
rint = ii #np.random.randint(1, 125)
simsubvol = lambda x: util.mask_func(x, rint)
sub_model.populate_mock(sub_halocat, masking_function=simsubvol, enforce_PBC=False)
pos = three_dim_pos_bundle(sub_model.mock.galaxy_table, 'x', 'y', 'z')
xi, yi , zi = util.random_shifter(rint)
temp_randoms = randoms.copy()
temp_randoms[:,0] += xi
temp_randoms[:,1] += yi
temp_randoms[:,2] += zi
rmax = xi_bin.max()
approx_cell1_size = [rmax , rmax , rmax]
approx_cellran_size = [rmax , rmax , rmax]
sub_xi = tpcf(
pos, xi_bin, pos,
randoms=temp_randoms,
period = None,
max_sample_size=int(1e5),
estimator=method,
approx_cell1_size = approx_cell1_size,
approx_cellran_size = approx_cellran_size,
RR_precomputed=RR,
NR_precomputed=NR)
label = None
if ii == N_sub - 1:
label = 'Subvolumes'
#if not ratio:
# sub.plot(0.5*(xi_bin[:-1]+xi_bin[1:]), sub_xi, lw=0.5, ls='--', c=pretty_colors[iii])
sub_xis += sub_xi
sub_xis_list.append(sub_xi)
sub_xi_avg = sub_xis/np.float(N_sub)
data_dump[method] = {}
data_dump[method]['sub_xi_avg'] = sub_xi_avg
data_dump[method]['sub_xis_list'] = sub_xis_list
else:
sub_xis_list = data_dump[method]['sub_xis_list']
sub_xi_avg = data_dump[method]['sub_xi_avg']
if not ratio:
sub.plot(0.5*(xi_bin[:-1]+xi_bin[1:]), sub_xi_avg,
lw=2, ls='--', c=pretty_colors[iii], label='Subvolume '+method)
else:
sub.plot(0.5*(xi_bin[:-1]+xi_bin[1:]), sub_xi_avg/full_xi,
lw=2, ls='--', c=pretty_colors[iii], label='Subvolume '+method)
if not os.path.isfile(pickle_file):
pickle.dump(data_dump, open(pickle_file, 'wb'))
sub.set_xlim([0.1, 50.])
sub.set_xlabel('r', fontsize=30)
sub.set_xscale('log')
if not ratio:
sub.set_ylabel(r"$\xi \mathtt{(r)}$", fontsize=25)
sub.set_yscale('log')
else:
sub.set_ylabel(r"$\overline{\xi^\mathtt{sub}}/\xi^\mathtt{all}$", fontsize=25)
sub.legend(loc='lower left')
if ratio:
fig_file = ''.join([util.fig_dir(), 'test_xi_subvolume_analytic.Nsub', str(N_sub), '.ratio.png'])
else:
fig_file = ''.join([util.fig_dir(), 'test_xi_subvolume_analytic.Nsub', str(N_sub), '.png'])
fig.savefig(fig_file, bbox_inches='tight', dpi=100)
plt.close()
return None
def build_nbar_xi_gmf(Mr=21, b_normal=0.25):
''' Build data vector [nbar, xi, gmf] and save to file
This data vector is built from the zeroth slice of the multidark
The other slices will be used for building the covariance matrix.
Parameters
----------
Mr : (int)
Absolute magnitude cut off M_r. Default M_r = -21.
b_normal : (float)
FoF Linking length
'''
thr = -1. * np.float(Mr)
model = PrebuiltHodModelFactory('zheng07', threshold=thr)
halocat = CachedHaloCatalog(simname = 'multidark', redshift = 0, halo_finder = 'rockstar')
####model.new_haloprop_func_dict = {'sim_subvol': util.mk_id_column}
####datsubvol = lambda x: util.mask_func(x, 0)
####model.populate_mock(halocat, masking_function=datsubvol, enforce_PBC=False)
model.populate_mock(halocat)
#all the things necessary for tpcf calculation
pos = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
#masking the galaxies outside the subvolume 0
pos = util.mask_galaxy_table(pos , 0)
rbins = xi_binedges()
rmax = rbins.max()
approx_cell1_size = [rmax , rmax , rmax]
approx_cellran_size = [rmax , rmax , rmax]
#compute number density
nbar = len(pos) / 200**3.
# load MD subvolume randoms and RRs
randoms = data_random(box='md_sub')
RR = data_RR(box='md_sub')
NR = len(randoms)
#compue tpcf with Natural estimator
data_xir = tpcf(
pos, rbins, pos,
randoms=randoms, period=None,
max_sample_size=int(2e5), estimator='Natural',
approx_cell1_size=approx_cell1_size,
approx_cellran_size=approx_cellran_size,
RR_precomputed=RR,
NR_precomputed=NR)
fullvec = np.append(nbar, data_xir)
#compute gmf
b = b_normal * (nbar)**(-1./3)
groups = pyfof.friends_of_friends(pos , b)
w = np.array([len(x) for x in groups])
gbins = gmf_bins()
gmf = np.histogram(w , gbins)[0] / (200.**3.)
fullvec = np.append(fullvec, gmf)
output_file = data_file(Mr=Mr, b_normal=b_normal)
np.savetxt(output_file, fullvec)
return None
def build_nbar_xi_gmf(Mr=21, b_normal=0.25):
''' Build data vector [nbar, xi, gmf] and save to file
This data vector is built from the zeroth slice of the multidark
The other slices will be used for building the covariance matrix.
Parameters
----------
Mr : (int)
Absolute magnitude cut off M_r. Default M_r = -21.
b_normal : (float)
FoF Linking length
'''
thr = -1. * np.float(Mr)
model = PrebuiltHodModelFactory('zheng07', threshold=thr)
halocat = CachedHaloCatalog(simname='multidark',
redshift=0,
halo_finder='rockstar')
####model.new_haloprop_func_dict = {'sim_subvol': util.mk_id_column}
####datsubvol = lambda x: util.mask_func(x, 0)
####model.populate_mock(halocat, masking_function=datsubvol, enforce_PBC=False)
model.populate_mock(halocat)
#all the things necessary for tpcf calculation
pos = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
#masking the galaxies outside the subvolume 0
pos = util.mask_galaxy_table(pos, 0)
rbins = xi_binedges()
rmax = rbins.max()
approx_cell1_size = [rmax, rmax, rmax]
approx_cellran_size = [rmax, rmax, rmax]
#compute number density
nbar = len(pos) / 200**3.
# load MD subvolume randoms and RRs
randoms = data_random(box='md_sub')
RR = data_RR(box='md_sub')
NR = len(randoms)
#compue tpcf with Natural estimator
data_xir = tpcf(pos,
rbins,
pos,
randoms=randoms,
period=None,
max_sample_size=int(2e5),
estimator='Natural',
approx_cell1_size=approx_cell1_size,
approx_cellran_size=approx_cellran_size,
RR_precomputed=RR,
NR_precomputed=NR)
fullvec = np.append(nbar, data_xir)
#compute gmf
b = b_normal * (nbar)**(-1. / 3)
groups = pyfof.friends_of_friends(pos, b)
w = np.array([len(x) for x in groups])
gbins = gmf_bins()
gmf = np.histogram(w, gbins)[0] / (200.**3.)
fullvec = np.append(fullvec, gmf)
output_file = data_file(Mr=Mr, b_normal=b_normal)
np.savetxt(output_file, fullvec)
return None
def build_MCMC_cov_nbar_xi_gmf(Mr=21, b_normal=0.25):
''' Build covariance matrix used in MCMC for the full nbar, xi, gmf data vector
using realisations of galaxy mocks for "data" HOD parameters in the
halos from the other subvolumes (subvolume 1 to subvolume 125) of
the simulation. Covariance matrices for different sets of observables
can be extracted from the full covariance matrix by slicing through
the indices.
'''
nbars = []
xir = []
gmfs = []
thr = -1. * np.float(Mr)
model = PrebuiltHodModelFactory('zheng07', threshold=thr)
halocat = CachedHaloCatalog(simname = 'multidark', redshift = 0, halo_finder = 'rockstar')
###model.new_haloprop_func_dict = {'sim_subvol': util.mk_id_column}
#some settings for tpcf calculations
rbins = xi_binedges()
rmax = rbins.max()
approx_cell1_size = [rmax , rmax , rmax]
approx_cellran_size = [rmax , rmax , rmax]
#load randoms and RRs
randoms = data_random(box='md_sub')
RR = data_RR(box='md_sub')
NR = len(randoms)
for i in xrange(1,125):
print 'mock#', i
# populate the mock subvolume
###mocksubvol = lambda x: util.mask_func(x, i)
###model.populate_mock(halocat,
### masking_function=mocksubvol,
### enforce_PBC=False)
model.populate_mock(halocat)
# returning the positions of galaxies in the entire volume
pos = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
# masking out the galaxies outside the subvolume i
pos = util.mask_galaxy_table(pos , i)
# calculate nbar
print "shape of pos" , pos.shape
nbars.append(len(pos) / 200**3.)
# translate the positions of randoms to the new subbox
xi0 , yi0 , zi0 = util.random_shifter(i)
temp_randoms = randoms.copy()
temp_randoms[:,0] += xi0
temp_randoms[:,1] += yi0
temp_randoms[:,2] += zi0
#calculate xi(r)
xi=tpcf(
pos, rbins, pos,
randoms=temp_randoms, period=None,
max_sample_size=int(3e5), estimator='Natural',
approx_cell1_size=approx_cell1_size,
approx_cellran_size=approx_cellran_size,
RR_precomputed = RR,
NR_precomputed = NR)
xir.append(xi)
# calculate gmf
nbar = len(pos) / 200**3.
b = b_normal * (nbar)**(-1./3)
groups = pyfof.friends_of_friends(pos , b)
w = np.array([len(x) for x in groups])
gbins = gmf_bins()
gmf = np.histogram(w , gbins)[0] / 200.**3.
gmfs.append(gmf)
# save nbar variance
nbar_var = np.var(nbars, axis=0, ddof=1)
nbar_file = ''.join([util.obvs_dir(), 'nbar_var.Mr', str(Mr), '.dat'])
np.savetxt(nbar_file, [nbar_var])
# write full covariance matrix of various combinations of the data
# and invert for the likelihood evaluations
# --- covariance for all three ---
fulldatarr = np.hstack((np.array(nbars).reshape(len(nbars), 1),
np.array(xir),
np.array(gmfs)))
fullcov = np.cov(fulldatarr.T)
fullcorr = np.corrcoef(fulldatarr.T)
# and save the covariance matrix
nopoisson_file = ''.join([util.obvs_dir(),
'MCMC.nbar_xi_gmf_cov', '.no_poisson', '.Mr', str(Mr), '.bnorm', str(round(b_normal,2)), '.dat'])
np.savetxt(nopoisson_file, fullcov)
return None
def build_ABC_cov_nbar_xi_gmf(Mr=21, b_normal=0.25):
''' Build covariance matrix used in ABC for the full nbar, xi, gmf data vector
using realisations of galaxy mocks for "data" HOD parameters in the
halos from the multidark simulation. Covariance matrices for different sets of observables
can be extracted from the full covariance matrix by slicing through
the indices.
Notes
-----
* This covariance matrix is the covariance matrix calculated from the *entire* multidark
box. So this does _not_ account for the sample variance, which the MCMC covariance does.
'''
nbars, xir, gmfs = [], [], []
thr = -1. * np.float(Mr)
model = PrebuiltHodModelFactory('zheng07', threshold=thr)
halocat = CachedHaloCatalog(simname='multidark',
redshift=0,
halo_finder='rockstar')
rbins = xi_binedges() # some setting for tpcf calculations
rmax = rbins.max()
approx_cell1_size = [rmax, rmax, rmax]
approx_cellran_size = [rmax, rmax, rmax]
# load randoms and RRs for the ENTIRE MultiDark volume
###randoms = data_random(box='md_all')
###RR = data_RR(box='md_all')
###NR = len(randoms)
for i in xrange(1, 125):
print 'mock#', i
# populate the mock subvolume
model.populate_mock(halocat)
# returning the positions of galaxies
pos = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
# calculate nbar
nbars.append(len(pos) / 1000**3.)
# calculate xi(r) for the ENTIRE MultiDark volume
# using the natural estimator DD/RR - 1
xi = tpcf(pos,
rbins,
period=model.mock.Lbox,
max_sample_size=int(3e5),
estimator='Natural',
approx_cell1_size=approx_cell1_size)
xir.append(xi)
# calculate gmf
nbar = len(pos) / 1000**3.
b = b_normal * (nbar)**(-1. / 3)
groups = pyfof.friends_of_friends(pos, b)
w = np.array([len(x) for x in groups])
gbins = gmf_bins()
gmf = np.histogram(w, gbins)[0] / (1000.**3.)
gmfs.append(gmf) # GMF
# save nbar variance
nbar_var = np.var(nbars, axis=0, ddof=1)
nbar_file = ''.join([util.obvs_dir(), 'abc_nbar_var.Mr', str(Mr), '.dat'])
np.savetxt(nbar_file, [nbar_var])
# write full covariance matrix of various combinations of the data
# and invert for the likelihood evaluations
# --- covariance for all three ---
fulldatarr = np.hstack(
(np.array(nbars).reshape(len(nbars),
1), np.array(xir), np.array(gmfs)))
fullcov = np.cov(fulldatarr.T)
fullcorr = np.corrcoef(fulldatarr.T)
# and save the covariance matrix
nopoisson_file = ''.join([
util.obvs_dir(), 'ABC.nbar_xi_gmf_cov', '.no_poisson', '.Mr',
str(Mr), '.bnorm',
str(round(b_normal, 2)), '.dat'
])
np.savetxt(nopoisson_file, fullcov)
return None
def build_ABC_cov_nbar_xi_gmf(Mr=21, b_normal=0.25):
''' Build covariance matrix used in ABC for the full nbar, xi, gmf data vector
using realisations of galaxy mocks for "data" HOD parameters in the
halos from the multidark simulation. Covariance matrices for different sets of observables
can be extracted from the full covariance matrix by slicing through
the indices.
Notes
-----
* This covariance matrix is the covariance matrix calculated from the *entire* multidark
box. So this does _not_ account for the sample variance, which the MCMC covariance does.
'''
nbars, xir, gmfs = [], [], []
thr = -1. * np.float(Mr)
model = PrebuiltHodModelFactory('zheng07', threshold=thr)
halocat = CachedHaloCatalog(simname = 'multidark', redshift = 0, halo_finder = 'rockstar')
rbins = xi_binedges() # some setting for tpcf calculations
rmax = rbins.max()
approx_cell1_size = [rmax , rmax , rmax]
approx_cellran_size = [rmax , rmax , rmax]
# load randoms and RRs for the ENTIRE MultiDark volume
###randoms = data_random(box='md_all')
###RR = data_RR(box='md_all')
###NR = len(randoms)
for i in xrange(1,125):
print 'mock#', i
# populate the mock subvolume
model.populate_mock(halocat)
# returning the positions of galaxies
pos = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
# calculate nbar
nbars.append(len(pos) / 1000**3.)
# calculate xi(r) for the ENTIRE MultiDark volume
# using the natural estimator DD/RR - 1
xi = tpcf(
pos, rbins, period=model.mock.Lbox,
max_sample_size=int(3e5), estimator='Natural',
approx_cell1_size=approx_cell1_size)
xir.append(xi)
# calculate gmf
nbar = len(pos) / 1000**3.
b = b_normal * (nbar)**(-1./3)
groups = pyfof.friends_of_friends(pos , b)
w = np.array([len(x) for x in groups])
gbins = gmf_bins()
gmf = np.histogram(w , gbins)[0] / (1000.**3.)
gmfs.append(gmf) # GMF
# save nbar variance
nbar_var = np.var(nbars, axis=0, ddof=1)
nbar_file = ''.join([util.obvs_dir(), 'abc_nbar_var.Mr', str(Mr), '.dat'])
np.savetxt(nbar_file, [nbar_var])
# write full covariance matrix of various combinations of the data
# and invert for the likelihood evaluations
# --- covariance for all three ---
fulldatarr = np.hstack((np.array(nbars).reshape(len(nbars), 1),
np.array(xir),
np.array(gmfs)))
fullcov = np.cov(fulldatarr.T)
fullcorr = np.corrcoef(fulldatarr.T)
# and save the covariance matrix
nopoisson_file = ''.join([util.obvs_dir(),
'ABC.nbar_xi_gmf_cov', '.no_poisson', '.Mr', str(Mr), '.bnorm', str(round(b_normal, 2)), '.dat'])
np.savetxt(nopoisson_file, fullcov)
return None
def main():
#cov = np.loadtxt("../../data/wpxicov_dr72_bright0_mr21.0_z0.159_nj400")
#f_MD = (1. + 71.74*10**6. / (1000.)**3.)
#f_bol = (1. + 71.74*10.**6. / (250.)**3.)
#print("covariance correction factor=" , f_bol/f_MD)
#cov = cov*f_bol/f_MD
gmf_cat = np.loadtxt("gmf20.dat")
data_gmf = gmf_cat[:,2]
data_gmf_err = (gmf_cat[:,3]**2. + gmf_cat[:,4]**2.)**.5
data_gmf_bin = np.hstack([gmf_cat[:,0],gmf_cat[-1,1]])
print(data_gmf_bin)
model.param_dict['logM0'] = 11.16
model.param_dict['sigma_logM'] = 0.61
model.param_dict['logMmin'] = 12.10
model.param_dict['alpha'] = 1.08
model.param_dict['logM1'] = 13.33
model.param_dict['mean_occupation_centrals_assembias_param1'] = -1.
halocat = CachedHaloCatalog(simname = 'bolplanck', redshift = 0, halo_finder = 'rockstar')
model.populate_mock(halocat)
pos = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
print("modelnumber density=" , len(pos)/halocat.Lbox**3.)
print("data number density=" , 1.16*10**-3.)
x = model.mock.galaxy_table['x']
y = model.mock.galaxy_table['y']
z = model.mock.galaxy_table['z']
vz = model.mock.galaxy_table['vz']
# applying RSD
pos = return_xyz_formatted_array(x, y, z, velocity = vz, velocity_distortion_dimension = 'z')
# enfprcing PBC
#pos = enforce_periodicity_of_box(pos, halocat.Lbox)
from halotools.mock_observables import FoFGroups
bperp = 0.14
bpar = 0.75
Lbox = np.array([halocat.Lbox,halocat.Lbox,halocat.Lbox])
period = Lbox
groups = FoFGroups(pos, b_perp=bperp, b_para=bpar, Lbox = Lbox , period=Lbox)
#model.mock.galaxy_table['fof_group_id'] = groups.group_ids
#model.mock.galaxy_table.sort(['fof_group_id'])
#from halotools.utils import group_member_generator
#grouping_key = 'fof_group_id'
#requested_columns = []
#group_gen = group_member_generator(model.mock.galaxy_table, grouping_key, requested_columns)
#group_richness = np.zeros(len(model.mock.galaxy_table), dtype=int)
#for first, last, member_props in group_gen:
# group_richness[first:last] = last-first
group_ids = groups.group_ids
group_richness = richness(group_ids)
gmf = np.histogram(np.array(group_richness) , data_gmf_bin)[0] / halocat.Lbox**3.
chi = np.sum(((gmf - data_gmf)**2. / data_gmf_err**2.)[:-1])
print(chi/(len(gmf)-1))
bin_centers = 0.5 * (data_gmf_bin[1:] + data_gmf_bin[:-1])
plt.figure(figsize=(10,10))
plt.errorbar(bin_centers , data_gmf , data_gmf_err , fmt=".k" , capsize = 2)
plt.plot(bin_centers , gmf)
plt.yscale("log")
plt.xlim([0,60])
plt.savefig("gmf.pdf")
def build_nbar_xi_gmf_cov(Mr=21):
''' Build covariance matrix for the full nbar, xi, gmf data vector
using realisations of galaxy mocks for "data" HOD parameters in the
halos from the multidark simulation. Covariance matrices for different sets of observables
can be extracted from the full covariance matrix by slicing through
the indices.
'''
nbars = []
xir = []
gmfs = []
thr = -1. * np.float(Mr)
model = PrebuiltHodModelFactory('zheng07', threshold=thr)
halocat = CachedHaloCatalog(simname = 'multidark', redshift = 0, halo_finder = 'rockstar')
#some settings for tpcf calculations
rbins = hardcoded_xi_bins()
for i in xrange(1,125):
print 'mock#', i
# populate the mock subvolume
model.populate_mock(halocat)
# returning the positions of galaxies
pos = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
# calculate nbar
nbars.append(len(pos) / 1000**3.)
# translate the positions of randoms to the new subbox
#calculate xi(r)
xi = tpcf(pos, rbins, period = model.mock.Lbox,
max_sample_size=int(2e5), estimator='Landy-Szalay')
xir.append(xi)
# calculate gmf
nbar = len(pos) / 1000**3.
b_normal = 0.75
b = b_normal * (nbar)**(-1./3)
groups = pyfof.friends_of_friends(pos , b)
w = np.array([len(x) for x in groups])
gbins = gmf_bins()
gmf = np.histogram(w , gbins)[0] / (1000.**3.)
gmfs.append(gmf) # GMF
# save nbar variance
nbar_var = np.var(nbars, axis=0, ddof=1)
nbar_file = ''.join([util.multidat_dir(), 'abc_nbar_var.Mr', str(Mr), '.dat'])
np.savetxt(nbar_file, [nbar_var])
# write full covariance matrix of various combinations of the data
# and invert for the likelihood evaluations
# --- covariance for all three ---
fulldatarr = np.hstack((np.array(nbars).reshape(len(nbars), 1),
np.array(xir),
np.array(gmfs)))
fullcov = np.cov(fulldatarr.T)
fullcorr = np.corrcoef(fulldatarr.T)
# and save the covariance matrix
nopoisson_file = ''.join([util.multidat_dir(), 'abc_nbar_xi_gmf_cov.no_poisson.Mr', str(Mr), '.dat'])
np.savetxt(nopoisson_file, fullcov)
# and a correlation matrix
full_corr_file = ''.join([util.multidat_dir(), 'abc_nbar_xi_gmf_corr.Mr', str(Mr), '.dat'])
np.savetxt(full_corr_file, fullcorr)
return None
def test_precomputed_rr(Nr, Mr=21):
'''
Mr = Luminositty threshold
Nr = Number of randoms
'''
rbins = np.logspace(-1, 1.25, 15)
rmax = rbins.max()
rbin_centers = (rbins[1:] + rbins[0:-1]) / 2.
halocat = CachedHaloCatalog(simname='bolshoi', redshift=0)
model = PrebuiltHodModelFactory("zheng07")
model.populate_mock(halocat=halocat, enforce_PBC=False)
data = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
print data.shape
L = halocat.Lbox
xmin, ymin, zmin = 0., 0., 0.
xmax, ymax, zmax = L, L, L
num_randoms = Nr
xran = np.random.uniform(xmin, xmax, num_randoms)
yran = np.random.uniform(ymin, ymax, num_randoms)
zran = np.random.uniform(zmin, zmax, num_randoms)
randoms = np.vstack((xran, yran, zran)).T
verbose = False
num_threads = cpu_count()
period = None
approx_cell1_size = [rmax, rmax, rmax]
approx_cell2_size = approx_cell1_size
approx_cellran_size = [rmax, rmax, rmax]
normal_result = tpcf(data,
rbins,
data,
randoms=randoms,
period=period,
max_sample_size=int(1e4),
estimator='Landy-Szalay',
approx_cell1_size=approx_cell1_size,
approx_cellran_size=approx_cellran_size)
#count data pairs
DD = npairs(data, data, rbins, period, verbose, num_threads,
approx_cell1_size, approx_cell2_size)
DD = np.diff(DD)
#count random pairs
RR = npairs(randoms, randoms, rbins, period, verbose, num_threads,
approx_cellran_size, approx_cellran_size)
RR = np.diff(RR)
#count data random pairs
DR = npairs(data, randoms, rbins, period, verbose, num_threads,
approx_cell1_size, approx_cell2_size)
DR = np.diff(DR)
print "DD=", DD
print "DR=", DR
print "RR=", RR
ND = len(data)
NR = len(randoms)
factor1 = ND * ND / (NR * NR)
factor2 = ND * NR / (NR * NR)
mult = lambda x, y: x * y
xi_LS = mult(1.0 / factor1, DD / RR) - mult(1.0 / factor2,
2.0 * DR / RR) + 1.0
print "xi=", xi_LS
print "normal=", normal_result
result_with_RR_precomputed = tpcf(data,
rbins,
data,
randoms=randoms,
period=period,
max_sample_size=int(1e5),
estimator='Landy-Szalay',
approx_cell1_size=approx_cell1_size,
approx_cellran_size=approx_cellran_size,
RR_precomputed=RR,
NR_precomputed=NR)
print "xi_pre=", result_with_RR_precomputed
def Subvolume_Analytic(N_sub, ratio=False):
''' Test the 2PCF estimates from MultiDark subvolume versus the
analytic 2PCF for the entire MultiDark volume
Parameters
----------
N_sub : (int)
Number of subvolumes to sample
'''
prettyplot()
pretty_colors = prettycolors()
pickle_file = ''.join([
'/export/bbq2/hahn/ccppabc/dump/', 'xi_subvolume_test', '.Nsub',
str(N_sub), '.p'
])
fig = plt.figure(1)
sub = fig.add_subplot(111)
xi_bin = xi_binedges()
if os.path.isfile(pickle_file):
data_dump = pickle.load(open(pickle_file, 'rb'))
full_xi = data_dump['full_xi']
else:
# Entire MultiDark Volume (Analytic xi)
model = PrebuiltHodModelFactory('zheng07', threshold=-21)
halocat = CachedHaloCatalog(simname='multidark',
redshift=0,
halo_finder='rockstar')
model.populate_mock(halocat)
pos = three_dim_pos_bundle(model.mock.galaxy_table, 'x', 'y', 'z')
# while the estimator claims to be Landy-Szalay, I highly suspect it
# actually uses Landy-Szalay since DR pairs cannot be calculated from
# analytic randoms
full_xi = tpcf(pos,
xi_bin,
period=model.mock.Lbox,
max_sample_size=int(2e5),
estimator='Landy-Szalay',
num_threads=1)
data_dump = {}
data_dump['full_xi'] = full_xi
if not ratio:
sub.plot(0.5 * (xi_bin[:-1] + xi_bin[1:]),
full_xi,
lw=2,
ls='-',
c='k',
label=r'Analytic $\xi$ Entire Volume')
if not os.path.isfile(pickle_file):
# MultiDark SubVolume (precomputed RR pairs)
sub_model = PrebuiltHodModelFactory('zheng07', threshold=-21)
sub_model.new_haloprop_func_dict = {'sim_subvol': util.mk_id_column}
sub_halocat = CachedHaloCatalog(simname='multidark',
redshift=0,
halo_finder='rockstar')
RR = data_RR()
randoms = data_random()
NR = len(randoms)
for method in ['Landy-Szalay', 'Natural']:
if method == 'Landy-Szalay':
iii = 3
elif method == 'Natural':
iii = 5
if not os.path.isfile(pickle_file):
sub_xis_list = []
sub_xis = np.zeros(len(full_xi))
for ii in range(1, N_sub + 1):
# randomly sample one of the subvolumes
rint = ii #np.random.randint(1, 125)
simsubvol = lambda x: util.mask_func(x, rint)
sub_model.populate_mock(sub_halocat,
masking_function=simsubvol,
enforce_PBC=False)
pos = three_dim_pos_bundle(sub_model.mock.galaxy_table, 'x',
'y', 'z')
xi, yi, zi = util.random_shifter(rint)
temp_randoms = randoms.copy()
temp_randoms[:, 0] += xi
temp_randoms[:, 1] += yi
temp_randoms[:, 2] += zi
rmax = xi_bin.max()
approx_cell1_size = [rmax, rmax, rmax]
approx_cellran_size = [rmax, rmax, rmax]
sub_xi = tpcf(pos,
xi_bin,
pos,
randoms=temp_randoms,
period=None,
max_sample_size=int(1e5),
estimator=method,
approx_cell1_size=approx_cell1_size,
approx_cellran_size=approx_cellran_size,
RR_precomputed=RR,
NR_precomputed=NR)
label = None
if ii == N_sub - 1:
label = 'Subvolumes'
#if not ratio:
# sub.plot(0.5*(xi_bin[:-1]+xi_bin[1:]), sub_xi, lw=0.5, ls='--', c=pretty_colors[iii])
sub_xis += sub_xi
sub_xis_list.append(sub_xi)
sub_xi_avg = sub_xis / np.float(N_sub)
data_dump[method] = {}
data_dump[method]['sub_xi_avg'] = sub_xi_avg
data_dump[method]['sub_xis_list'] = sub_xis_list
else:
sub_xis_list = data_dump[method]['sub_xis_list']
sub_xi_avg = data_dump[method]['sub_xi_avg']
if not ratio:
sub.plot(0.5 * (xi_bin[:-1] + xi_bin[1:]),
sub_xi_avg,
lw=2,
ls='--',
c=pretty_colors[iii],
label='Subvolume ' + method)
else:
sub.plot(0.5 * (xi_bin[:-1] + xi_bin[1:]),
sub_xi_avg / full_xi,
lw=2,
ls='--',
c=pretty_colors[iii],
label='Subvolume ' + method)
if not os.path.isfile(pickle_file):
pickle.dump(data_dump, open(pickle_file, 'wb'))
sub.set_xlim([0.1, 50.])
sub.set_xlabel('r', fontsize=30)
sub.set_xscale('log')
if not ratio:
sub.set_ylabel(r"$\xi \mathtt{(r)}$", fontsize=25)
sub.set_yscale('log')
else:
sub.set_ylabel(r"$\overline{\xi^\mathtt{sub}}/\xi^\mathtt{all}$",
fontsize=25)
sub.legend(loc='lower left')
if ratio:
fig_file = ''.join([
util.fig_dir(), 'test_xi_subvolume_analytic.Nsub',
str(N_sub), '.ratio.png'
])
else:
fig_file = ''.join([
util.fig_dir(), 'test_xi_subvolume_analytic.Nsub',
str(N_sub), '.png'
])
fig.savefig(fig_file, bbox_inches='tight', dpi=100)
plt.close()
return None
