-
Notifications
You must be signed in to change notification settings - Fork 0
/
GOSE_TTV_extractor.py
186 lines (151 loc) · 7.42 KB
/
GOSE_TTV_extractor.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
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
from __future__ import division
import numpy as np
import matplotlib.pyplot as plt
import pandas
import os
import time
import traceback
#### this script will use the GOSE posteriors to produce a giant list (a la Holczer 2016) of transit timings.
"""
For each planet, do *joint posterior* draws for each segment, grab those timings, and use the median value across that segment.
Repeat for each segment.
The, fit a line to the timings -- that's your linear ephemeris.
Then you can compute O - C values from this.
CHECK YOUR LINEAR EPHEMERIS CALCULATION AGAINST THE FIDUCIAL VALUE.
Question: should you fit lines to each segment individually? Or one line to all of them? (I think the lines should be fit individually, but you should test this -- visualize it)
"""
try:
GOSE_TTV_summaryfile = open('/data/tethys/Documents/Central_Data/GOSE_TTV_summaryfile.csv', mode='w')
GOSE_TTV_summaryfile.write('KOI,Pplan_pub,Pplan_fit,epoch,tau_linear_ephemeris,tau_fit,OC_min,OCmin_err\n')
GOSE_TTV_summaryfile.close()
show_debug_plots = input("Do you want to show plots for debugging? y/n: ")
TTVpostdir = '/run/media/amteachey/Auddy_Akiti/Teachey/Nmoon_TTVs/GOSE_TTVs/TTV_posteriors-master'
TTVpostfiles = np.array(os.listdir(TTVpostdir))
TTVpostfiles_copy = []
for TTVpostfile in TTVpostfiles:
if '.csv' in TTVpostfile:
TTVpostfiles_copy.append(TTVpostfile)
TTVpostfiles = np.array(TTVpostfiles_copy)
### sort them by KOI
KOI_nums = []
for TTVpostfile in TTVpostfiles:
KOI_nums.append(TTVpostfile[4:TTVpostfile.find('_')])
KOI_nums = np.array(KOI_nums).astype(float)
KOI_nums_argsort = np.argsort(KOI_nums)
TTVpostfiles = TTVpostfiles[KOI_nums_argsort]
ndraws = 500
for nTTVpostfile, TTVpostfile in enumerate(TTVpostfiles):
if '.csv' not in TTVpostfile:
continue
KOI = TTVpostfile[:TTVpostfile.find('_')] ### filename is something like KOI-41.03_TTV_posteriors.csv, so this should isolate the KOI name.
print(KOI)
try:
TTVpost = pandas.read_csv(TTVpostdir+'/'+TTVpostfile)
except:
print('there was a problem opening this file.')
time.sleep(5)
continue
TTVpostcols = TTVpost.columns
column_prefixes = []
for TTVpostcol in TTVpostcols:
TTVpostcol_prefix = TTVpostcol[:TTVpostcol.find('_')] ### will isolate seg0, seg1, ... seg11, etc.
column_prefixes.append(TTVpostcol_prefix)
column_prefixes = np.array(column_prefixes)
unique_column_prefixes = np.unique(column_prefixes)
nsegs = len(unique_column_prefixes)
try:
for seg in np.arange(0,nsegs,1):
seg_RpRstar_post = np.array(TTVpost['seg'+str(seg)+'_p'])
seg_rhostar_post = np.array(TTVpost['seg'+str(seg)+'_rhostar'])
seg_impact_post = np.array(TTVpost['seg'+str(seg)+'_b'])
seg_pplan_post = np.array(TTVpost['seg'+str(seg)+'_Pp'])
seg_tauref_post = np.array(TTVpost['seg'+str(seg)+'_tau_ref'])
seg_q1_post = np.array(TTVpost['seg'+str(seg)+'_q1'])
seg_q2_post = np.array(TTVpost['seg'+str(seg)+'_q2'])
seg_ntaus = 0
seg_tau_dict = {}
for tau in np.arange(0,20,1):
try:
seg_tau_dict[tau] = np.array(TTVpost['seg'+str(seg)+'_tau'+str(tau)])
except:
break
seg_likelihood_post = np.array(TTVpost['seg'+str(seg)+'_likelihood'])
#### now we want to pull out ndraws-worth of indices
seg_draw_idxs = np.random.randint(low=0, high=len(seg_RpRstar_post), size=ndraws)
seg_median_pplan = np.nanmedian(seg_pplan_post[seg_draw_idxs])
seg_median_tauref = np.nanmedian(seg_tauref_post[seg_draw_idxs])
seg_median_taus = []
seg_median_tau_stds = []
for tau in seg_tau_dict.keys():
seg_median_taus.append(np.nanmedian(seg_tau_dict[tau][seg_draw_idxs]))
seg_median_tau_stds.append(np.nanstd(seg_tau_dict[tau][seg_draw_idxs]))
seg_median_taus = np.array(seg_median_taus)
seg_median_tau_stds = np.array(seg_median_tau_stds)
#### now we have all the transit timings... want to fit a line to these.
###### CAREFUL! THESE ARE NOT NECESSARILY UNIFORMLY SPACED!
#### COMPUTE THE EPOCH NUMBER BASED ON IT'S DISTANCE FROM TAUREF AND THE SEGMENT PERIOD!
###### for example: if tauref = 0, period = 35, and tau1 = 71, then let (tau1 - tauref) // 35 == 2. EPOCH 2.
#seg_median_tau_epochs = (seg_median_taus - seg_median_tauref) // seg_median_pplan
seg_median_tau_epochs = np.around(((seg_median_taus - seg_median_tauref) / seg_median_pplan), 0)
if show_debug_plots == 'y':
print("seg_median_pplan = ", seg_median_pplan)
print('seg_median_tauref = ', seg_median_tauref)
print('seg_median_taus = ', seg_median_taus)
print(' ')
print('seg_median_tau_epochs ', seg_median_tau_epochs)
#### the problem is, there aren't enough transit timings here to make a good period prediction is some cases.
###### so what you really want to do is add all these epochs together, and fit at the very end.
if seg == 0:
koi_epochs = seg_median_tau_epochs
koi_taus = seg_median_taus
koi_tau_stds = seg_median_tau_stds
else:
#### append to lists above
koi_epochs = np.concatenate((koi_epochs, seg_median_tau_epochs))
koi_taus = np.concatenate((koi_taus, seg_median_taus))
koi_tau_stds = np.concatenate((koi_tau_stds, seg_median_tau_stds))
if len(koi_epochs) < 2:
continue
bad_tau_idxs = np.where(koi_taus < 0)[0]
koi_epochs, koi_taus, koi_tau_stds = np.delete(koi_epochs, bad_tau_idxs), np.delete(koi_taus, bad_tau_idxs), np.delete(koi_tau_stds, bad_tau_idxs)
koi_linfit_coefs = np.polyfit(x=koi_epochs, y=koi_taus, deg=1, w=1/koi_tau_stds)
### sanity check the coefficients
koi_linfit_slope, koi_linfit_intercept = koi_linfit_coefs
print('nsegs = ', nsegs)
print('koi_linfit_slope, expected pplan = ', koi_linfit_slope, seg_median_pplan)
koi_linfit_function = np.poly1d(koi_linfit_coefs)
### now compute the line value at each epoch
koi_linear_ephemeris_timings = koi_linfit_function(koi_epochs)
OminusC_days = koi_taus - koi_linear_ephemeris_timings
OminusC_minutes = OminusC_days * 24 * 60
OminusC_minutes_errors = koi_tau_stds * 24 * 60
if show_debug_plots == 'y':
"""
plt.scatter(koi_epochs, koi_taus, facecolor='LightCoral', edgecolor='k', s=20, zorder=1)
plt.errorbar(koi_epochs, koi_taus, yerr=koi_tau_stds, ecolor='k', zorder=0, fmt='none')
plt.plot(koi_epochs, koi_linear_ephemeris_timings, linestyle=':', color='k')
plt.xlabel('Epoch')
#plt.ylabel(r'$\Tau$')
plt.show()
"""
### OK IT WORKS
plt.scatter(koi_epochs, OminusC_minutes, facecolor='LightCoral', edgecolor='k', s=20, zorder=1)
plt.errorbar(koi_epochs, OminusC_minutes, yerr=OminusC_minutes_errors, ecolor='k', zorder=0, fmt='none')
plt.plot(koi_epochs, np.linspace(0,0,len(koi_epochs)), color='k', linestyle=':')
plt.xlabel('Epoch')
plt.ylabel('O - C [min]')
plt.title(KOI)
plt.show()
print(' ')
GOSE_TTV_summaryfile = open('/data/tethys/Documents/Central_Data/GOSE_TTV_summaryfile.csv', mode='a')
print('writing to file.')
for epoch, koi_epoch in enumerate(koi_epochs):
#GOSE_TTV_summaryfile.write('KOI,Pplan_pub,Pplan_fit,epoch,tau_expected,tau_fit,OC_min,OCmin_err\n')
GOSE_TTV_summaryfile.write(str(KOI)+','+str(seg_median_pplan)+','+str(koi_linfit_slope)+','+str(koi_epoch)+','+str(koi_linear_ephemeris_timings[epoch])+','+str(koi_taus[epoch])+','+str(OminusC_minutes[epoch])+','+str(OminusC_minutes_errors[epoch])+'\n')
GOSE_TTV_summaryfile.close()
except:
print("unable to open this planet file. skipping.")
time.sleep(5)
continue
except:
traceback.print_exc()