-
Notifications
You must be signed in to change notification settings - Fork 1
/
bias_r_pk.py
132 lines (108 loc) · 4.4 KB
/
bias_r_pk.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
import sys
import glob
import numpy as np
from scipy import interpolate
from stats import chi2
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
from matplotlib import gridspec
import mpl_style
plt.style.use(mpl_style.style1)
Testing = False
zz = 0.987
ngrid = 128 # The one used in pkg.py
sims = ['UNITSIM1']#,'UNITSIM1_InvPhase','UNITSIM2','UNITSIM2_InvPhase']
lboxes = [1000.]*len(sims) # Mpc/h
# Limits to obtain the bias
kminb = 0.01
kmaxb = 0.1
############################################
seldir = '/home2/vgonzalez/out/desi_samUNIT/'
############################################
if Testing: sims = [sims[0]]
# Read theoretical linear P(k,zz)
thfil = seldir+'linPk_z'+str(zz).replace('.','_')+'.dat'
k_th,plin_th,pnl_th = np.loadtxt(thfil,unpack=True)
fpth = interpolate.interp1d(k_th,plin_th)
# Array for trying out bias
abias = np.linspace(0.1,20.,10000)
# Read the galaxy P(k,zz)
for iis,sim in enumerate(sims):
inpath = seldir+sim
knyquist = np.pi*ngrid/lboxes[iis]
volumen = lboxes[iis]**3
# Plot for r-space
figr = plt.figure()
gsr = gridspec.GridSpec(3,1) ; gsr.update(wspace=0., hspace=0.)
cm = plt.get_cmap('tab10') # Colour map to draw colours from
cols = ['k','k']
axr = plt.subplot(gsr[2,0]) # Ratio plot
axr.set_xlabel("$k$ [$h$/Mpc]")
axr.set_ylabel("$\\sqrt{P_{\\rm i}/P_{\\rm DM}}$")
axr.set_autoscale_on(False) ; axr.minorticks_on()
axr.set_xlim(0.01,knyquist) ; axr.set_ylim(0.8,2)
axr.set_xscale('log')
axr.plot(k_th,plin_th/plin_th,color=cols[0])
axr.plot(k_th,pnl_th/plin_th,color=cols[1],linestyle='--')
axp = plt.subplot(gsr[0:2,0],sharex=axr) # Pk plot
plt.setp(axp.get_xticklabels(), visible=False)
axp.set_autoscale_on(False) ; axp.minorticks_on()
axp.set_yscale('log') ; axp.set_ylim(11,100000.)
axp.set_ylabel('P($k$) [Mpc/$h$)$^3$]')
axp.axvline(x=knyquist,color=cols[1],linestyle=':')
axp.plot(k_th,plin_th,color=cols[0],label='DM, z='+str(zz))
axp.plot(k_th,pnl_th,color=cols[1],linestyle='--',label='DM, NL')
# r-space
files = sorted(glob.glob(inpath+'/ascii_files/Pk/Pk*dat'))
if Testing: files = [files[0]]
bfile = inpath+'/ascii_files/Pk/bias_kmin'+str(kminb).replace('.','_')+\
'_kmax'+str(kmaxb).replace('.','_')+\
'_z'+str(zz).replace('.','_')+'.dat'
with open(bfile,'w') as outf:
outf.write('# bias selection \n')
for ii,ff in enumerate(files):
col = cm(1.*ii/len(files))
cols.append(col)
# Get number density
root = ff.split('/Pk/Pk_')[1].split('_kmin')[0]
galff = ff.split('Pk/')[0] + root +'.dat'
xgal = np.loadtxt(galff,usecols=(0),unpack=True)
nd = len(xgal)/volumen ; xgal = []
# Read the power spectrum and calculate the corresponding error
kg,pkg = np.loadtxt(ff,unpack=True)
dk = kg[1] - kg[0] # Sim. bin
errorPk = np.sqrt(((2*np.pi)**2/(kg**2*dk*volumen))*(pkg + 1/nd)**2)
pth = fpth(kg)
# Bias calculation r-space
chis = np.zeros((len(abias))) ; chis.fill(999.) ; bias = -999.
ind = np.where((kg>=kminb) & (kg<kmaxb))
if (np.shape(ind)[1]>1):
for ib,bb in enumerate(abias):
obs = pkg[ind]
model = bb*bb*pth[ind]
error = errorPk[ind]
chis[ib] = chi2(obs,model,error)
ib = np.where(chis == np.nanmin(chis))
bias = abias[ib][0]
with open(bfile,'a') as outf:
outf.write(str(bias)+' '+ff.split('/Pk/Pk_')[1]+' \n')
print('nd={:.3f} for file {}, bias={}'.format(np.log10(nd),root,bias))
# Plot bias r-space
axr.plot(kg,np.sqrt(pkg/pth),color=col)
axr.axhline(bias,color=col,linestyle=':')
# Plot bias P(k)
label = ff.split('Pk_')[1].split('_z')[0]
axp.fill_between(kg,pkg-errorPk,pkg+errorPk,color=col,alpha=0.4)
axp.plot(kg,pkg,color=col,label=label)
axp.axhline(1/nd,color=col,linestyle=':')
# Legend
leg = axp.legend(loc=0,fontsize='small')
leg.draw_frame(False)
for ii,text in enumerate(leg.get_texts()):
text.set_color(cols[ii])
# Save figure
plotfile = inpath+'/plots/Pk_rspace_z'+str(zz).replace('.','_')+'.pdf'
figr.savefig(plotfile,constrained_layout=True)
print('Output: ',plotfile)
print(' r-space bias in ',bfile)