Cheetah/lcsinglefit.py at master · lmwalkowicz/Cheetah · GitHub

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
187
188

from numpy import *
from scipy.optimize import leastsq
from lcspot import *
from scipy.cluster.vq import whiten
from scipy.cluster.vq import kmeans2
import sys
if '-p' in sys.argv: import matplotlib.pyplot as plt


def vincfit(lctuple, p0s=[], initsteps=20, nclusters=30, threshratio=2, plsprint='some', plsplot=False):
  phase, intensity = lctuple

  if plsprint != 'none':
    print 'first spot'
    print 'lightcurve:'
    print intensity
    print

  #initial fits using leastsq, starting at various points in param space
  opts,sses,p0s = vilcfits(phase,intensity,p0s,initsteps)

  #sort by sses
  stups = sorted(zip(sses,opts,p0s), key=lambda x: x[0])
  sses,opts,p0s = zip(*stups)

  #cluster best fits
  optsmat = whiten(array(opts[0:3*len(opts)/4]))
  centroid,label = kmeans2(optsmat,nclusters,iter=20,minit='points')
  ll = list(label)
  newp0s = [opts[ll.index(i)] for i in range(0,nclusters) if i in ll]

  #full fits
  opts,sses,p1s = vilcfits(phase,intensity,newp0s)

  #sort by sses again
  stups = sorted(zip(sses,opts,p1s), key=lambda x: x[0])
  sses,opts,p1s = zip(*stups)

  #find best fit(s)
  threshold = sses[0]*threshratio
  bestps = [opts[0]]
  bestsses = [sses[0]]
  bestp1s = [p1s[0]]
  for i in range(1,len(opts)):
    if sses[i] > threshold:
      break
    if min([paramdist(opts[i],k) for k in bestps]) > 0.001:
      bestps.append(opts[i])
      bestsses.append(sses[i])
      bestp1s.append(p1s[i])

  #print best fit(s)
  if plsprint != 'none':
    print 'best fits:'
    for i in range(0,len(bestps)):
      print 'opt:',bestps[i],'sse:',bestsses[i],'p1:',bestp1s[i]
    print

  #print all fits
  if plsprint == 'all':
    print 'all fits:'
    for i in range(0,len(opts)):
      print 'opt:',opts[i],'sse:',sses[i],'p1:',p1s[i]
    print

  #plot
  if plsplot:
    colors = 'bgcmyk'
    plt.figure()
    plt.plot(phase, intensity, 'r.')
    for i in range(0,len(bestps)):
      plt.plot(phase, lcspot(phase,bestps[i]), colors[i%len(colors)])
    plt.title('initial single spot fit')
    plt.xlabel('phase')
    plt.ylabel('intensity')
    #plt.show()

  return bestps


def fincfit(lctuple, inc, p0s=[], initsteps=20, nclusters=20, threshratio=2, plsprint='some', plsplot=False, snum=-1):
  phase, intensity = lctuple

  if plsprint != 'none':
    print 'inc:', inc, 'spot #', snum
    print 'residual lightcurve:'
    print intensity
    print

  #initial fits using leastsq, starting at various points in param space
  opts,sses,p0s = filcfits(phase,intensity,inc,p0s,initsteps)

  #sort by sses
  stups = sorted(zip(sses,opts,p0s), key=lambda x: x[0])
  sses,opts,p0s = zip(*stups)

  #cluster best fits
  optsmat = whiten(array(opts[0:3*len(opts)/4]))
  centroid,label = kmeans2(optsmat,nclusters,iter=20,minit='points')
  ll = list(label)
  newp0s = [opts[ll.index(i)] for i in range(0,nclusters) if i in ll]

  #full fits
  opts,sses,p1s = filcfits(phase,intensity,inc,newp0s)

  #sort by sses again
  stups = sorted(zip(sses,opts,p1s), key=lambda x: x[0])
  sses,opts,p1s = zip(*stups)

  #find best fit(s)
  threshold = sses[0]*threshratio
  bestps = [opts[0]]
  bestsses = [sses[0]]
  bestp1s = [p1s[0]]
  for i in range(1,len(opts)):
    if sses[i] > threshold:
      break
    if min([spotparamdist(opts[i],k) for k in bestps]) > 0.001:
      bestps.append(opts[i])
      bestsses.append(sses[i])
      bestp1s.append(p1s[i])

  #print best fit(s)
  if plsprint != 'none':
    print 'best fits:'
    for i in range(0,len(bestps)):
      print 'opt:',bestps[i],'sse:',bestsses[i],'p1:',bestp1s[i]
    print

  #print all fits
  if plsprint == 'all':
    print 'all fits:'
    for i in range(0,len(opts)):
      print 'opt:',opts[i],'sse:',sses[i],'p1:',p1s[i]
    print

  #plot
  if plsplot:
    colors = 'bgcmyk'
    plt.figure()
    plt.plot(phase, intensity, 'r.')
    for i in range(0,len(bestps)):
      plt.plot(phase, lcspotfi(phase,inc,bestps[i]), colors[i%len(colors)])
    plt.xlabel('phase')
    plt.ylabel('residual intensity')
    plt.title('inc: ' + str(inc) + ', spot #' + str(snum))
    #plt.show()

  return bestps


def vilcfits(phase,intensity,p0s=[],mfev=0):
  if not p0s:
    p0s = spacedparams(5)
  opts = []
  sses = []
  p0s2 = []
  for p0 in p0s:
    try:
      fps, iem = leastsq(lcspotdiffs3, p0, args=(phase,intensity), maxfev=mfev)
      #fps[1] = fps[1] % 360.0
      #fps[3] = abs(fps[3])
      opts.append(fps)
      sses.append(lcspotsse(fps,phase,intensity))
      p0s2.append(p0)
    except RuntimeError as e:
      print "Runtime Error({0}): {1}".format(e.errno, e.strerror)
  return opts,sses,p0s2


def filcfits(phase,intensity,inc,p0s=[],mfev=0):
  if not p0s:
    p0s = spacedparamsfi(5)
  opts = []
  sses = []
  p0s2 = []
  for p0 in p0s:
    try:
      fps, iem = leastsq(lcspotdiffsfi, p0, args=(phase,intensity,inc), maxfev=mfev)
      #fps[0] = fps[1] % 360.0
      #fps[2] = abs(fps[2])
      opts.append(fps)
      sses.append(lcspotssefi(fps,phase,intensity,inc))
      p0s2.append(p0)
    except RuntimeError as e:
      print "Runtime Error({0}): {1}".format(e.errno, e.strerror)
  return opts,sses,p0s2