pyfstat.py 72.5 KB
Newer Older
1
2
3
4
5
6
7
8
9
10
""" Classes for various types of searches using ComputeFstatistic """
import os
import sys
import itertools
import logging
import argparse
import copy
import glob
import inspect
from functools import wraps
11
import subprocess
12
from collections import OrderedDict
13
14
15

import numpy as np
import matplotlib
16
matplotlib.use('Agg')
17
18
19
20
import matplotlib.pyplot as plt
import emcee
import corner
import dill as pickle
21
import lal
22
23
import lalpulsar

24
plt.rcParams['text.usetex'] = True
25
plt.rcParams['axes.formatter.useoffset'] = False
26

27
28
29
30
31
32
33
config_file = os.path.expanduser('~')+'/.pyfstat.conf'
if os.path.isfile(config_file):
    d = {}
    with open(config_file, 'r') as f:
        for line in f:
            k, v = line.split('=')
            k = k.replace(' ', '')
34
            v = v.replace(' ', '').replace("'", "").replace('"', '').replace('\n', '')
35
36
37
38
39
40
            d[k] = v
    earth_ephem = d['earth_ephem']
    sun_ephem = d['sun_ephem']
else:
    logging.warning('No ~/.pyfstat.conf file found please provide the paths '
                    'when initialising searches')
41
42
43
    earth_ephem = None
    sun_ephem = None

44
45
46
47
48
parser = argparse.ArgumentParser()
parser.add_argument("-q", "--quite", help="Decrease output verbosity",
                    action="store_true")
parser.add_argument("-c", "--clean", help="Don't use cached data",
                    action="store_true")
49
parser.add_argument("-u", "--use-old-data", action="store_true")
50
51
52
53
54
55
56
57
58
59
60
61
62
parser.add_argument('unittest_args', nargs='*')
args, unknown = parser.parse_known_args()
sys.argv[1:] = args.unittest_args

if args.quite:
    log_level = logging.WARNING
else:
    log_level = logging.DEBUG

logging.basicConfig(level=log_level,
                    format='%(asctime)s %(levelname)-8s: %(message)s',
                    datefmt='%H:%M')

63
64

def initializer(func):
65
    """ Automatically assigns the parameters to self """
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
    names, varargs, keywords, defaults = inspect.getargspec(func)

    @wraps(func)
    def wrapper(self, *args, **kargs):
        for name, arg in list(zip(names[1:], args)) + list(kargs.items()):
            setattr(self, name, arg)

        for name, default in zip(reversed(names), reversed(defaults)):
            if not hasattr(self, name):
                setattr(self, name, default)

        func(self, *args, **kargs)

    return wrapper


def read_par(label, outdir):
83
    """ Read in a .par file, returns a dictionary of the values """
84
85
86
87
    filename = '{}/{}.par'.format(outdir, label)
    d = {}
    with open(filename, 'r') as f:
        for line in f:
88
89
90
            if len(line.split('=')) > 1:
                key, val = line.rstrip('\n').split(' = ')
                key = key.strip()
91
                d[key] = np.float64(eval(val.rstrip('; ')))
92
93
94
95
    return d


class BaseSearchClass(object):
96
    """ The base search class, provides ephemeris and general utilities """
97
98
99
100

    earth_ephem_default = earth_ephem
    sun_ephem_default = sun_ephem

101
102
103
104
105
106
107
108
109
    def add_log_file(self):
        ' Log output to a log-file, requires class to have outdir and label '
        logfilename = '{}/{}.log'.format(self.outdir, self.label)
        fh = logging.FileHandler(logfilename)
        fh.setFormatter(logging.Formatter(
            '%(asctime)s %(levelname)-8s: %(message)s',
            datefmt='%y-%m-%d %H:%M'))
        logging.getLogger().addHandler(fh)

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
    def shift_matrix(self, n, dT):
        """ Generate the shift matrix """
        m = np.zeros((n, n))
        factorial = np.math.factorial
        for i in range(n):
            for j in range(n):
                if i == j:
                    m[i, j] = 1.0
                elif i > j:
                    m[i, j] = 0.0
                else:
                    if i == 0:
                        m[i, j] = 2*np.pi*float(dT)**(j-i) / factorial(j-i)
                    else:
                        m[i, j] = float(dT)**(j-i) / factorial(j-i)

        return m

    def shift_coefficients(self, theta, dT):
        """ Shift a set of coefficients by dT

        Parameters
        ----------
        theta: array-like, shape (n,)
            vector of the expansion coefficients to transform starting from the
135
            lowest degree e.g [phi, F0, F1,...].
136
        dT: float
137
            difference between the two reference times as tref_new - tref_old.
138
139
140
141

        Returns
        -------
        theta_new: array-like shape (n,)
142
            vector of the coefficients as evaluate as the new reference time.
143
144
145
146
147
        """
        n = len(theta)
        m = self.shift_matrix(n, dT)
        return np.dot(m, theta)

148
    def calculate_thetas(self, theta, delta_thetas, tbounds, theta0_idx=0):
149
150
151
        """ Calculates the set of coefficients for the post-glitch signal """
        thetas = [theta]
        for i, dt in enumerate(delta_thetas):
152
153
154
155
156
157
158
159
160
161
162
163
164
            if i < theta0_idx:
                pre_theta_at_ith_glitch = self.shift_coefficients(
                    thetas[0], tbounds[i+1] - self.tref)
                post_theta_at_ith_glitch = pre_theta_at_ith_glitch - dt
                thetas.insert(0, self.shift_coefficients(
                    post_theta_at_ith_glitch, self.tref - tbounds[i+1]))

            elif i >= theta0_idx:
                pre_theta_at_ith_glitch = self.shift_coefficients(
                    thetas[i], tbounds[i+1] - self.tref)
                post_theta_at_ith_glitch = pre_theta_at_ith_glitch + dt
                thetas.append(self.shift_coefficients(
                    post_theta_at_ith_glitch, self.tref - tbounds[i+1]))
165
166
167
        return thetas


Gregory Ashton's avatar
Gregory Ashton committed
168
169
170
171
172
173
174
class ComputeFstat(object):
    """ Base class providing interface to lalpulsar.ComputeFstat """

    earth_ephem_default = earth_ephem
    sun_ephem_default = sun_ephem

    @initializer
175
    def __init__(self, tref, sftfilepath=None,
176
                 minStartTime=None, maxStartTime=None,
Gregory Ashton's avatar
Gregory Ashton committed
177
                 minCoverFreq=None, maxCoverFreq=None,
178
                 detector=None, earth_ephem=None, sun_ephem=None,
179
                 binary=False, transient=True, BSGL=False):
180
181
182
183
184
        """
        Parameters
        ----------
        tref: int
            GPS seconds of the reference time.
185
186
        sftfilepath: str
            File patern to match SFTs
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
        minCoverFreq, maxCoverFreq: float
            The min and max cover frequency passed to CreateFstatInput, if
            either is None the range of frequencies in the SFT less 1Hz is
            used.
        detector: str
            Two character reference to the data to use, specify None for no
            contraint.
        earth_ephem, sun_ephem: str
            Paths of the two files containing positions of Earth and Sun,
            respectively at evenly spaced times, as passed to CreateFstatInput.
            If None defaults defined in BaseSearchClass will be used.
        binary: bool
            If true, search of binary parameters.
        transient: bool
            If true, allow for the Fstat to be computed over a transient range.
202
203
        BSGL: bool
            If true, compute the BSGL rather than the twoF value.
204
205

        """
Gregory Ashton's avatar
Gregory Ashton committed
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220

        if earth_ephem is None:
            self.earth_ephem = self.earth_ephem_default
        if sun_ephem is None:
            self.sun_ephem = self.sun_ephem_default

        self.init_computefstatistic_single_point()

    def init_computefstatistic_single_point(self):
        """ Initilisation step of run_computefstatistic for a single point """

        logging.info('Initialising SFTCatalog')
        constraints = lalpulsar.SFTConstraints()
        if self.detector:
            constraints.detector = self.detector
221
222
223
224
225
        if self.minStartTime:
            constraints.minStartTime = lal.LIGOTimeGPS(self.minStartTime)
        if self.maxStartTime:
            constraints.maxStartTime = lal.LIGOTimeGPS(self.maxStartTime)

226
        logging.info('Loading data matching pattern {}'.format(
227
228
                     self.sftfilepath))
        SFTCatalog = lalpulsar.SFTdataFind(self.sftfilepath, constraints)
Gregory Ashton's avatar
Gregory Ashton committed
229
        names = list(set([d.header.name for d in SFTCatalog.data]))
230
        epochs = [d.header.epoch for d in SFTCatalog.data]
231
        logging.info(
232
233
            'Loaded {} data files from detectors {} spanning {} to {}'.format(
                len(epochs), names, int(epochs[0]), int(epochs[-1])))
Gregory Ashton's avatar
Gregory Ashton committed
234
235
236
237
238
239

        logging.info('Initialising ephems')
        ephems = lalpulsar.InitBarycenter(self.earth_ephem, self.sun_ephem)

        logging.info('Initialising FstatInput')
        dFreq = 0
240
241
242
243
244
        if self.transient:
            self.whatToCompute = lalpulsar.FSTATQ_ATOMS_PER_DET
        else:
            self.whatToCompute = lalpulsar.FSTATQ_2F

Gregory Ashton's avatar
Gregory Ashton committed
245
246
247
248
249
250
251
252
        FstatOptionalArgs = lalpulsar.FstatOptionalArgsDefaults

        if self.minCoverFreq is None or self.maxCoverFreq is None:
            fA = SFTCatalog.data[0].header.f0
            numBins = SFTCatalog.data[0].numBins
            fB = fA + (numBins-1)*SFTCatalog.data[0].header.deltaF
            self.minCoverFreq = fA + 0.5
            self.maxCoverFreq = fB - 0.5
253
254
255
            logging.info('Min/max cover freqs not provided, using '
                         '{} and {}, est. from SFTs'.format(
                             self.minCoverFreq, self.maxCoverFreq))
Gregory Ashton's avatar
Gregory Ashton committed
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275

        self.FstatInput = lalpulsar.CreateFstatInput(SFTCatalog,
                                                     self.minCoverFreq,
                                                     self.maxCoverFreq,
                                                     dFreq,
                                                     ephems,
                                                     FstatOptionalArgs
                                                     )

        logging.info('Initialising PulsarDoplerParams')
        PulsarDopplerParams = lalpulsar.PulsarDopplerParams()
        PulsarDopplerParams.refTime = self.tref
        PulsarDopplerParams.Alpha = 1
        PulsarDopplerParams.Delta = 1
        PulsarDopplerParams.fkdot = np.array([0, 0, 0, 0, 0, 0, 0])
        self.PulsarDopplerParams = PulsarDopplerParams

        logging.info('Initialising FstatResults')
        self.FstatResults = lalpulsar.FstatResults()

276
277
278
279
        if self.BSGL:
            logging.info('Initialising BSGL: this will fail if numDet < 2')
            # Tuning parameters - to be reviewed
            numDetectors = 2
Gregory Ashton's avatar
Gregory Ashton committed
280
            Fstar0sc = 15.
281
            oLGX = np.zeros(10)
Gregory Ashton's avatar
Gregory Ashton committed
282
            oLGX[:numDetectors] = 1./numDetectors
283
284
285
286
287
288
289
290
291
            self.BSGLSetup = lalpulsar.CreateBSGLSetup(numDetectors,
                                                       Fstar0sc,
                                                       oLGX,
                                                       False,
                                                       1)
            self.twoFX = np.zeros(10)
            self.whatToCompute = (lalpulsar.FSTATQ_2F +
                                  lalpulsar.FSTATQ_2F_PER_DET)

292
        if self.transient:
293
            logging.info('Initialising transient parameters')
294
295
296
297
298
299
            self.windowRange = lalpulsar.transientWindowRange_t()
            self.windowRange.type = lalpulsar.TRANSIENT_RECTANGULAR
            self.windowRange.t0Band = 0
            self.windowRange.dt0 = 1
            self.windowRange.tauBand = 0
            self.windowRange.dtau = 1
300

Gregory Ashton's avatar
Gregory Ashton committed
301
    def run_computefstatistic_single_point(self, tstart, tend, F0, F1,
302
303
304
                                           F2, Alpha, Delta, asini=None,
                                           period=None, ecc=None, tp=None,
                                           argp=None):
305
        """ Returns the twoF fully-coherently at a single point """
Gregory Ashton's avatar
Gregory Ashton committed
306
307
308
309

        self.PulsarDopplerParams.fkdot = np.array([F0, F1, F2, 0, 0, 0, 0])
        self.PulsarDopplerParams.Alpha = Alpha
        self.PulsarDopplerParams.Delta = Delta
310
311
312
313
314
315
        if self.binary:
            self.PulsarDopplerParams.asini = asini
            self.PulsarDopplerParams.period = period
            self.PulsarDopplerParams.ecc = ecc
            self.PulsarDopplerParams.tp = tp
            self.PulsarDopplerParams.argp = argp
Gregory Ashton's avatar
Gregory Ashton committed
316
317
318
319

        lalpulsar.ComputeFstat(self.FstatResults,
                               self.FstatInput,
                               self.PulsarDopplerParams,
320
                               1,
Gregory Ashton's avatar
Gregory Ashton committed
321
322
323
                               self.whatToCompute
                               )

324
        if self.transient is False:
325
326
327
328
329
330
331
332
333
            if self.BSGL is False:
                return self.FstatResults.twoF[0]

            twoF = np.float(self.FstatResults.twoF[0])
            self.twoFX[0] = self.FstatResults.twoFPerDet(0)
            self.twoFX[1] = self.FstatResults.twoFPerDet(1)
            BSGL = lalpulsar.ComputeBSGL(twoF, self.twoFX,
                                         self.BSGLSetup)
            return BSGL
334

335
336
        self.windowRange.t0 = int(tstart)  # TYPE UINT4
        self.windowRange.tau = int(tend - tstart)  # TYPE UINT4
337

Gregory Ashton's avatar
Gregory Ashton committed
338
        FS = lalpulsar.ComputeTransientFstatMap(
339
            self.FstatResults.multiFatoms[0], self.windowRange, False)
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358

        if self.BSGL is False:
            return 2*FS.F_mn.data[0][0]

        FstatResults_single = copy.copy(self.FstatResults)
        FstatResults_single.lenth = 1
        FstatResults_single.data = self.FstatResults.multiFatoms[0].data[0]
        FS0 = lalpulsar.ComputeTransientFstatMap(
            FstatResults_single.multiFatoms[0], self.windowRange, False)
        FstatResults_single.data = self.FstatResults.multiFatoms[0].data[1]
        FS1 = lalpulsar.ComputeTransientFstatMap(
            FstatResults_single.multiFatoms[0], self.windowRange, False)

        self.twoFX[0] = 2*FS0.F_mn.data[0][0]
        self.twoFX[1] = 2*FS1.F_mn.data[0][0]
        BSGL = lalpulsar.ComputeBSGL(2*FS.F_mn.data[0][0], self.twoFX,
                                     self.BSGLSetup)

        return BSGL
Gregory Ashton's avatar
Gregory Ashton committed
359
360
361


class SemiCoherentGlitchSearch(BaseSearchClass, ComputeFstat):
362
363
364
365
366
367
368
369
370
    """ A semi-coherent glitch search

    This implements a basic `semi-coherent glitch F-stat in which the data
    is divided into two segments either side of the proposed glitch and the
    fully-coherent F-stat in each segment is averaged to give the semi-coherent
    F-stat
    """

    @initializer
Gregory Ashton's avatar
Gregory Ashton committed
371
    def __init__(self, label, outdir, tref, tstart, tend, nglitch=0,
372
                 sftfilepath=None, theta0_idx=0, BSGL=False,
373
374
375
                 minCoverFreq=None, maxCoverFreq=None, minStartTime=None,
                 maxStartTime=None, detector=None, earth_ephem=None,
                 sun_ephem=None):
376
377
378
379
        """
        Parameters
        ----------
        label, outdir: str
380
381
382
383
384
385
            A label and directory to read/write data from/to.
        tref, tstart, tend: int
            GPS seconds of the reference time, and start and end of the data.
        nglitch: int
            The (fixed) number of glitches; this can zero, but occasionally
            this causes issue (in which case just use ComputeFstat).
386
387
        sftfilepath: str
            File patern to match SFTs
388
389
390
391
        theta0_idx, int
            Index (zero-based) of which segment the theta refers to - uyseful
            if providing a tight prior on theta to allow the signal to jump
            too theta (and not just from)
392
        minCoverFreq, maxCoverFreq: float
393
394
395
            The min and max cover frequency passed to CreateFstatInput, if
            either is None the range of frequencies in the SFT less 1Hz is
            used.
396
397
        detector: str
            Two character reference to the data to use, specify None for no
398
            contraint.
399
400
        earth_ephem, sun_ephem: str
            Paths of the two files containing positions of Earth and Sun,
401
402
            respectively at evenly spaced times, as passed to CreateFstatInput.
            If None defaults defined in BaseSearchClass will be used.
403
404
405
406
407
408
409
        """

        self.fs_file_name = "{}/{}_FS.dat".format(self.outdir, self.label)
        if self.earth_ephem is None:
            self.earth_ephem = self.earth_ephem_default
        if self.sun_ephem is None:
            self.sun_ephem = self.sun_ephem_default
410
411
        self.transient = True
        self.binary = False
412
413
414
        self.init_computefstatistic_single_point()

    def compute_nglitch_fstat(self, F0, F1, F2, Alpha, Delta, *args):
415
        """ Returns the semi-coherent glitch summed twoF """
416
417
418

        args = list(args)
        tboundaries = [self.tstart] + args[-self.nglitch:] + [self.tend]
419
420
421
422
423
424
425
426
        delta_F0s = args[-3*self.nglitch:-2*self.nglitch]
        delta_F1s = args[-2*self.nglitch:-self.nglitch]
        delta_F2 = np.zeros(len(delta_F0s))
        delta_phi = np.zeros(len(delta_F0s))
        theta = [0, F0, F1, F2]
        delta_thetas = np.atleast_2d(
                np.array([delta_phi, delta_F0s, delta_F1s, delta_F2]).T)

427
428
        thetas = self.calculate_thetas(theta, delta_thetas, tboundaries,
                                       theta0_idx=self.theta0_idx)
429
430

        twoFSum = 0
431
        for i, theta_i_at_tref in enumerate(thetas):
432
433
434
            ts, te = tboundaries[i], tboundaries[i+1]

            twoFVal = self.run_computefstatistic_single_point(
435
436
                ts, te, theta_i_at_tref[1], theta_i_at_tref[2],
                theta_i_at_tref[3], Alpha, Delta)
437
438
            twoFSum += twoFVal

439
440
441
        if np.isfinite(twoFSum):
            return twoFSum
        else:
442
            return -np.inf
443
444
445

    def compute_glitch_fstat_single(self, F0, F1, F2, Alpha, Delta, delta_F0,
                                    delta_F1, tglitch):
446
447
448
449
        """ Returns the semi-coherent glitch summed twoF for nglitch=1

        Note: used for testing
        """
450
451
452
453
454
455
456
457
458
459
460

        theta = [F0, F1, F2]
        delta_theta = [delta_F0, delta_F1, 0]
        tref = self.tref

        theta_at_glitch = self.shift_coefficients(theta, tglitch - tref)
        theta_post_glitch_at_glitch = theta_at_glitch + delta_theta
        theta_post_glitch = self.shift_coefficients(
            theta_post_glitch_at_glitch, tref - tglitch)

        twoFsegA = self.run_computefstatistic_single_point(
Gregory Ashton's avatar
Gregory Ashton committed
461
            self.tstart, tglitch, theta[0], theta[1], theta[2], Alpha,
462
463
464
465
466
467
            Delta)

        if tglitch == self.tend:
            return twoFsegA

        twoFsegB = self.run_computefstatistic_single_point(
Gregory Ashton's avatar
Gregory Ashton committed
468
            tglitch, self.tend, theta_post_glitch[0],
469
470
471
472
473
474
            theta_post_glitch[1], theta_post_glitch[2], Alpha,
            Delta)

        return twoFsegA + twoFsegB


Gregory Ashton's avatar
Gregory Ashton committed
475
476
class MCMCSearch(BaseSearchClass):
    """ MCMC search using ComputeFstat"""
477
    @initializer
478
    def __init__(self, label, outdir, sftfilepath, theta_prior, tref,
479
                 tstart, tend, nsteps=[100, 100, 100], nwalkers=100, ntemps=1,
480
                 log10temperature_min=-5, theta_initial=None, scatter_val=1e-4,
481
482
                 binary=False, BSGL=False, minCoverFreq=None, maxCoverFreq=None,
                 detector=None, earth_ephem=None, sun_ephem=None, theta0_idx=0):
483
484
485
486
        """
        Parameters
        label, outdir: str
            A label and directory to read/write data from/to
487
488
        sftfilepath: str
            File patern to match SFTs
489
        theta_prior: dict
490
491
492
493
            Dictionary of priors and fixed values for the search parameters.
            For each parameters (key of the dict), if it is to be held fixed
            the value should be the constant float, if it is be searched, the
            value should be a dictionary of the prior.
494
495
496
497
        theta_initial: dict, array, (None)
            Either a dictionary of distribution about which to distribute the
            initial walkers about, an array (from which the walkers will be
            scattered by scatter_val, or  None in which case the prior is used.
498
499
500
501
502
503
504
        tref, tstart, tend: int
            GPS seconds of the reference time, start time and end time
        nsteps: list (m,)
            List specifying the number of steps to take, the last two entries
            give the nburn and nprod of the 'production' run, all entries
            before are for iterative initialisation steps (usually just one)
            e.g. [1000, 1000, 500].
505
506
507
508
509
510
511
512
513
514
515
        nwalkers, ntemps: int,
            The number of walkers and temperates to use in the parallel
            tempered PTSampler.
        log10temperature_min float < 0
            The  log_10(tmin) value, the set of betas passed to PTSampler are
            generated from np.logspace(0, log10temperature_min, ntemps).
        binary: Bool
            If true, search over binary parameters
        detector: str
            Two character reference to the data to use, specify None for no
            contraint.
516
517
518
519
520
521
522
523
524
525
        minCoverFreq, maxCoverFreq: float
            Minimum and maximum instantaneous frequency which will be covered
            over the SFT time span as passed to CreateFstatInput
        earth_ephem, sun_ephem: str
            Paths of the two files containing positions of Earth and Sun,
            respectively at evenly spaced times, as passed to CreateFstatInput
            If None defaults defined in BaseSearchClass will be used

        """

526
527
528
        self.minStartTime = tstart
        self.maxStartTime = tend

529
        self.add_log_file()
Gregory Ashton's avatar
Gregory Ashton committed
530
531
        logging.info(
            'Set-up MCMC search for model {} on data {}'.format(
532
                self.label, self.sftfilepath))
533
534
535
        if os.path.isdir(outdir) is False:
            os.mkdir(outdir)
        self.pickle_path = '{}/{}_saved_data.p'.format(self.outdir, self.label)
Gregory Ashton's avatar
Gregory Ashton committed
536
537
        self.theta_prior['tstart'] = self.tstart
        self.theta_prior['tend'] = self.tend
538
539
        self.unpack_input_theta()
        self.ndim = len(self.theta_keys)
540
541
542
543
        if self.log10temperature_min:
            self.betas = np.logspace(0, self.log10temperature_min, self.ntemps)
        else:
            self.betas = None
544

545
546
547
548
549
550
551
552
553
        if earth_ephem is None:
            self.earth_ephem = self.earth_ephem_default
        if sun_ephem is None:
            self.sun_ephem = self.sun_ephem_default

        if args.clean and os.path.isfile(self.pickle_path):
            os.rename(self.pickle_path, self.pickle_path+".old")

        self.old_data_is_okay_to_use = self.check_old_data_is_okay_to_use()
554
555
556
        self.log_input()

    def log_input(self):
557
        logging.info('theta_prior = {}'.format(self.theta_prior))
558
        logging.info('nwalkers={}'.format(self.nwalkers))
559
560
561
562
        logging.info('scatter_val = {}'.format(self.scatter_val))
        logging.info('nsteps = {}'.format(self.nsteps))
        logging.info('ntemps = {}'.format(self.ntemps))
        logging.info('log10temperature_min = {}'.format(
563
            self.log10temperature_min))
564
565
566

    def inititate_search_object(self):
        logging.info('Setting up search object')
Gregory Ashton's avatar
Gregory Ashton committed
567
        self.search = ComputeFstat(
568
569
570
571
            tref=self.tref, sftfilepath=self.sftfilepath,
            minCoverFreq=self.minCoverFreq, maxCoverFreq=self.maxCoverFreq,
            earth_ephem=self.earth_ephem, sun_ephem=self.sun_ephem,
            detector=self.detector, BSGL=self.BSGL, transient=False,
572
            minStartTime=self.minStartTime, maxStartTime=self.maxStartTime)
573
574

    def logp(self, theta_vals, theta_prior, theta_keys, search):
Gregory Ashton's avatar
Gregory Ashton committed
575
        H = [self.generic_lnprior(**theta_prior[key])(p) for p, key in
576
577
578
579
580
581
             zip(theta_vals, theta_keys)]
        return np.sum(H)

    def logl(self, theta, search):
        for j, theta_i in enumerate(self.theta_idxs):
            self.fixed_theta[theta_i] = theta[j]
Gregory Ashton's avatar
Gregory Ashton committed
582
        FS = search.run_computefstatistic_single_point(*self.fixed_theta)
583
584
585
        return FS

    def unpack_input_theta(self):
Gregory Ashton's avatar
Gregory Ashton committed
586
587
        full_theta_keys = ['tstart', 'tend', 'F0', 'F1', 'F2', 'Alpha',
                           'Delta']
588
589
590
        if self.binary:
            full_theta_keys += [
                'asini', 'period', 'ecc', 'tp', 'argp']
591
592
        full_theta_keys_copy = copy.copy(full_theta_keys)

Gregory Ashton's avatar
Gregory Ashton committed
593
594
        full_theta_symbols = ['_', '_', '$f$', '$\dot{f}$', '$\ddot{f}$',
                              r'$\alpha$', r'$\delta$']
595
596
597
598
        if self.binary:
            full_theta_symbols += [
                'asini', 'period', 'period', 'ecc', 'tp', 'argp']

599
600
        self.theta_keys = []
        fixed_theta_dict = {}
601
        for key, val in self.theta_prior.iteritems():
602
603
            if type(val) is dict:
                fixed_theta_dict[key] = 0
Gregory Ashton's avatar
Gregory Ashton committed
604
                self.theta_keys.append(key)
605
606
607
608
609
610
            elif type(val) in [float, int, np.float64]:
                fixed_theta_dict[key] = val
            else:
                raise ValueError(
                    'Type {} of {} in theta not recognised'.format(
                        type(val), key))
Gregory Ashton's avatar
Gregory Ashton committed
611
            full_theta_keys_copy.pop(full_theta_keys_copy.index(key))
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627

        if len(full_theta_keys_copy) > 0:
            raise ValueError(('Input dictionary `theta` is missing the'
                              'following keys: {}').format(
                                  full_theta_keys_copy))

        self.fixed_theta = [fixed_theta_dict[key] for key in full_theta_keys]
        self.theta_idxs = [full_theta_keys.index(k) for k in self.theta_keys]
        self.theta_symbols = [full_theta_symbols[i] for i in self.theta_idxs]

        idxs = np.argsort(self.theta_idxs)
        self.theta_idxs = [self.theta_idxs[i] for i in idxs]
        self.theta_symbols = [self.theta_symbols[i] for i in idxs]
        self.theta_keys = [self.theta_keys[i] for i in idxs]

    def check_initial_points(self, p0):
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
        for nt in range(self.ntemps):
            logging.info('Checking temperature {} chains'.format(nt))
            initial_priors = np.array([
                self.logp(p, self.theta_prior, self.theta_keys, self.search)
                for p in p0[nt]])
            number_of_initial_out_of_bounds = sum(initial_priors == -np.inf)

            if number_of_initial_out_of_bounds > 0:
                logging.warning(
                    'Of {} initial values, {} are -np.inf due to the prior'
                    .format(len(initial_priors),
                            number_of_initial_out_of_bounds))

                p0 = self.generate_new_p0_to_fix_initial_points(
                    p0, nt, initial_priors)

    def generate_new_p0_to_fix_initial_points(self, p0, nt, initial_priors):
        logging.info('Attempting to correct intial values')
        idxs = np.arange(self.nwalkers)[initial_priors == -np.inf]
        count = 0
        while sum(initial_priors == -np.inf) > 0 and count < 100:
            for j in idxs:
                p0[nt][j] = (p0[nt][np.random.randint(0, self.nwalkers)]*(
                             1+np.random.normal(0, 1e-10, self.ndim)))
            initial_priors = np.array([
                self.logp(p, self.theta_prior, self.theta_keys,
                          self.search)
                for p in p0[nt]])
            count += 1

        if sum(initial_priors == -np.inf) > 0:
            logging.info('Failed to fix initial priors')
        else:
            logging.info('Suceeded to fix initial priors')

        return p0
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683

    def run(self):

        if self.old_data_is_okay_to_use is True:
            logging.warning('Using saved data from {}'.format(
                self.pickle_path))
            d = self.get_saved_data()
            self.sampler = d['sampler']
            self.samples = d['samples']
            self.lnprobs = d['lnprobs']
            self.lnlikes = d['lnlikes']
            return

        self.inititate_search_object()

        sampler = emcee.PTSampler(
            self.ntemps, self.nwalkers, self.ndim, self.logl, self.logp,
            logpargs=(self.theta_prior, self.theta_keys, self.search),
            loglargs=(self.search,), betas=self.betas)

Gregory Ashton's avatar
Gregory Ashton committed
684
685
        p0 = self.generate_initial_p0()
        p0 = self.apply_corrections_to_p0(p0)
686
687
688
689
690
        self.check_initial_points(p0)

        ninit_steps = len(self.nsteps) - 2
        for j, n in enumerate(self.nsteps[:-2]):
            logging.info('Running {}/{} initialisation with {} steps'.format(
691
                j+1, ninit_steps, n))
692
            sampler.run_mcmc(p0, n)
693
694
            logging.info("Mean acceptance fraction: {0:.3f}"
                         .format(np.mean(sampler.acceptance_fraction)))
695
696
697
            if self.ntemps > 1:
                logging.info("Tswap acceptance fraction: {}"
                             .format(sampler.tswap_acceptance_fraction))
Gregory Ashton's avatar
Gregory Ashton committed
698
            fig, axes = self.plot_walkers(sampler, symbols=self.theta_symbols)
699
700
701
            fig.savefig('{}/{}_init_{}_walkers.png'.format(
                self.outdir, self.label, j))

702
            p0 = self.get_new_p0(sampler)
Gregory Ashton's avatar
Gregory Ashton committed
703
            p0 = self.apply_corrections_to_p0(p0)
704
705
706
707
708
709
710
711
            self.check_initial_points(p0)
            sampler.reset()

        nburn = self.nsteps[-2]
        nprod = self.nsteps[-1]
        logging.info('Running final burn and prod with {} steps'.format(
            nburn+nprod))
        sampler.run_mcmc(p0, nburn+nprod)
712
713
        logging.info("Mean acceptance fraction: {0:.3f}"
                     .format(np.mean(sampler.acceptance_fraction)))
714
715
716
        if self.ntemps > 1:
            logging.info("Tswap acceptance fraction: {}"
                         .format(sampler.tswap_acceptance_fraction))
717

Gregory Ashton's avatar
Gregory Ashton committed
718
        fig, axes = self.plot_walkers(sampler, symbols=self.theta_symbols)
719
720
721
722
723
724
725
726
727
728
729
        fig.savefig('{}/{}_walkers.png'.format(self.outdir, self.label))

        samples = sampler.chain[0, :, nburn:, :].reshape((-1, self.ndim))
        lnprobs = sampler.lnprobability[0, :, nburn:].reshape((-1))
        lnlikes = sampler.lnlikelihood[0, :, nburn:].reshape((-1))
        self.sampler = sampler
        self.samples = samples
        self.lnprobs = lnprobs
        self.lnlikes = lnlikes
        self.save_data(sampler, samples, lnprobs, lnlikes)

730
    def plot_corner(self, figsize=(7, 7),  tglitch_ratio=False,
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
                    add_prior=False, nstds=None, label_offset=0.4,
                    dpi=300, rc_context={}, **kwargs):

        with plt.rc_context(rc_context):
            fig, axes = plt.subplots(self.ndim, self.ndim,
                                     figsize=figsize)

            samples_plt = copy.copy(self.samples)
            theta_symbols_plt = copy.copy(self.theta_symbols)
            theta_symbols_plt = [s.replace('_{glitch}', r'_\textrm{glitch}') for s
                                 in theta_symbols_plt]

            if tglitch_ratio:
                for j, k in enumerate(self.theta_keys):
                    if k == 'tglitch':
                        s = samples_plt[:, j]
                        samples_plt[:, j] = (s - self.tstart)/(
                                             self.tend - self.tstart)
                        theta_symbols_plt[j] = r'$R_{\textrm{glitch}}$'

            if type(nstds) is int and 'range' not in kwargs:
                _range = []
                for j, s in enumerate(samples_plt.T):
                    median = np.median(s)
                    std = np.std(s)
                    _range.append((median - nstds*std, median + nstds*std))
            else:
                _range = None

            fig_triangle = corner.corner(samples_plt,
                                         labels=theta_symbols_plt,
                                         fig=fig,
                                         bins=50,
                                         max_n_ticks=4,
                                         plot_contours=True,
                                         plot_datapoints=True,
                                         label_kwargs={'fontsize': 8},
                                         data_kwargs={'alpha': 0.1,
                                                      'ms': 0.5},
                                         range=_range,
                                         **kwargs)

            axes_list = fig_triangle.get_axes()
            axes = np.array(axes_list).reshape(self.ndim, self.ndim)
            plt.draw()
            for ax in axes[:, 0]:
                ax.yaxis.set_label_coords(-label_offset, 0.5)
            for ax in axes[-1, :]:
                ax.xaxis.set_label_coords(0.5, -label_offset)
            for ax in axes_list:
                ax.set_rasterized(True)
                ax.set_rasterization_zorder(-10)
            plt.tight_layout(h_pad=0.0, w_pad=0.0)
            fig.subplots_adjust(hspace=0.05, wspace=0.05)

            if add_prior:
                self.add_prior_to_corner(axes, samples_plt)

            fig_triangle.savefig('{}/{}_corner.png'.format(
                self.outdir, self.label), dpi=dpi)
791
792
793
794
795
796

    def add_prior_to_corner(self, axes, samples):
        for i, key in enumerate(self.theta_keys):
            ax = axes[i][i]
            xlim = ax.get_xlim()
            s = samples[:, i]
Gregory Ashton's avatar
Gregory Ashton committed
797
            prior = self.generic_lnprior(**self.theta_prior[key])
798
799
800
801
802
803
            x = np.linspace(s.min(), s.max(), 100)
            ax2 = ax.twinx()
            ax2.get_yaxis().set_visible(False)
            ax2.plot(x, [prior(xi) for xi in x], '-r')
            ax.set_xlim(xlim)

Gregory Ashton's avatar
Gregory Ashton committed
804
    def generic_lnprior(self, **kwargs):
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
        """ Return a lambda function of the pdf

        Parameters
        ----------
        kwargs: dict
            A dictionary containing 'type' of pdf and shape parameters

        """

        def logunif(x, a, b):
            above = x < b
            below = x > a
            if type(above) is not np.ndarray:
                if above and below:
                    return -np.log(b-a)
                else:
                    return -np.inf
            else:
                idxs = np.array([all(tup) for tup in zip(above, below)])
                p = np.zeros(len(x)) - np.inf
                p[idxs] = -np.log(b-a)
                return p

        def halfnorm(x, loc, scale):
            if x < 0:
                return -np.inf
            else:
                return -0.5*((x-loc)**2/scale**2+np.log(0.5*np.pi*scale**2))

        def cauchy(x, x0, gamma):
            return 1.0/(np.pi*gamma*(1+((x-x0)/gamma)**2))

        def exp(x, x0, gamma):
            if x > x0:
                return np.log(gamma) - gamma*(x - x0)
            else:
                return -np.inf

        if kwargs['type'] == 'unif':
            return lambda x: logunif(x, kwargs['lower'], kwargs['upper'])
        elif kwargs['type'] == 'halfnorm':
            return lambda x: halfnorm(x, kwargs['loc'], kwargs['scale'])
847
848
        elif kwargs['type'] == 'neghalfnorm':
            return lambda x: halfnorm(-x, kwargs['loc'], kwargs['scale'])
849
850
851
852
853
854
855
        elif kwargs['type'] == 'norm':
            return lambda x: -0.5*((x - kwargs['loc'])**2/kwargs['scale']**2
                                   + np.log(2*np.pi*kwargs['scale']**2))
        else:
            logging.info("kwargs:", kwargs)
            raise ValueError("Print unrecognise distribution")

Gregory Ashton's avatar
Gregory Ashton committed
856
    def generate_rv(self, **kwargs):
857
858
859
860
861
862
863
864
        dist_type = kwargs.pop('type')
        if dist_type == "unif":
            return np.random.uniform(low=kwargs['lower'], high=kwargs['upper'])
        if dist_type == "norm":
            return np.random.normal(loc=kwargs['loc'], scale=kwargs['scale'])
        if dist_type == "halfnorm":
            return np.abs(np.random.normal(loc=kwargs['loc'],
                                           scale=kwargs['scale']))
865
866
867
        if dist_type == "neghalfnorm":
            return -1 * np.abs(np.random.normal(loc=kwargs['loc'],
                                                scale=kwargs['scale']))
868
869
870
871
872
873
        if dist_type == "lognorm":
            return np.random.lognormal(
                mean=kwargs['loc'], sigma=kwargs['scale'])
        else:
            raise ValueError("dist_type {} unknown".format(dist_type))

Gregory Ashton's avatar
Gregory Ashton committed
874
875
    def plot_walkers(self, sampler, symbols=None, alpha=0.4, color="k", temp=0,
                     start=None, stop=None, draw_vline=None):
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
        """ Plot all the chains from a sampler """

        shape = sampler.chain.shape
        if len(shape) == 3:
            nwalkers, nsteps, ndim = shape
            chain = sampler.chain[:, :, :]
        if len(shape) == 4:
            ntemps, nwalkers, nsteps, ndim = shape
            if temp < ntemps:
                logging.info("Plotting temperature {} chains".format(temp))
            else:
                raise ValueError(("Requested temperature {} outside of"
                                  "available range").format(temp))
            chain = sampler.chain[temp, :, :, :]

        with plt.style.context(('classic')):
            fig, axes = plt.subplots(ndim, 1, sharex=True, figsize=(8, 4*ndim))

            if ndim > 1:
                for i in range(ndim):
896
                    axes[i].ticklabel_format(useOffset=False, axis='y')
897
898
                    cs = chain[:, start:stop, i].T
                    axes[i].plot(cs, color="k", alpha=alpha)
899
900
901
902
                    if symbols:
                        axes[i].set_ylabel(symbols[i])
                    if draw_vline is not None:
                        axes[i].axvline(draw_vline, lw=2, ls="--")
903
904
905
906
907

            else:
                cs = chain[:, start:stop, 0].T
                axes.plot(cs, color='k', alpha=alpha)
                axes.ticklabel_format(useOffset=False, axis='y')
908
909
910

        return fig, axes

Gregory Ashton's avatar
Gregory Ashton committed
911
912
913
914
915
    def apply_corrections_to_p0(self, p0):
        """ Apply any correction to the initial p0 values """
        return p0

    def generate_scattered_p0(self, p):
916
        """ Generate a set of p0s scattered about p """
Gregory Ashton's avatar
Gregory Ashton committed
917
        p0 = [[p + self.scatter_val * p * np.random.randn(self.ndim)
918
919
920
921
               for i in xrange(self.nwalkers)]
              for j in xrange(self.ntemps)]
        return p0

Gregory Ashton's avatar
Gregory Ashton committed
922
    def generate_initial_p0(self):
923
924
925
        """ Generate a set of init vals for the walkers """

        if type(self.theta_initial) == dict:
926
927
928
            logging.info('Generate initial values from initial dictionary')
            if self.nglitch > 1:
                raise ValueError('Initial dict not implemented for nglitch>1')
Gregory Ashton's avatar
Gregory Ashton committed
929
            p0 = [[[self.generate_rv(**self.theta_initial[key])
930
931
932
                    for key in self.theta_keys]
                   for i in range(self.nwalkers)]
                  for j in range(self.ntemps)]
933
934
935
936
937
938
        elif type(self.theta_initial) == list:
            logging.info('Generate initial values from list of theta_initial')
            p0 = [[[self.generate_rv(**val)
                    for val in self.theta_initial]
                   for i in range(self.nwalkers)]
                  for j in range(self.ntemps)]
939
        elif self.theta_initial is None:
940
            logging.info('Generate initial values from prior dictionary')
Gregory Ashton's avatar
Gregory Ashton committed
941
            p0 = [[[self.generate_rv(**self.theta_prior[key])
942
943
944
945
                    for key in self.theta_keys]
                   for i in range(self.nwalkers)]
                  for j in range(self.ntemps)]
        elif len(self.theta_initial) == self.ndim:
Gregory Ashton's avatar
Gregory Ashton committed
946
            p0 = self.generate_scattered_p0(self.theta_initial)
947
948
949
950
951
        else:
            raise ValueError('theta_initial not understood')

        return p0

952
    def get_new_p0(self, sampler):
953
954
955
956
957
958
959
960
961
962
963
964
        """ Returns new initial positions for walkers are burn0 stage

        This returns new positions for all walkers by scattering points about
        the maximum posterior with scale `scatter_val`.

        """
        if sampler.chain[:, :, -1, :].shape[0] == 1:
            ntemps_temp = 1
        else:
            ntemps_temp = self.ntemps
        pF = sampler.chain[:, :, -1, :].reshape(
            ntemps_temp, self.nwalkers, self.ndim)[0, :, :]
965
966
        lnl = sampler.lnlikelihood[:, :, -1].reshape(
            self.ntemps, self.nwalkers)[0, :]
967
968
        lnp = sampler.lnprobability[:, :, -1].reshape(
            self.ntemps, self.nwalkers)[0, :]
969
970

        # General warnings about the state of lnp
971
        if any(np.isnan(lnp)):
972
973
974
975
976
977
978
979
980
981
982
            logging.warning(
                "Of {} lnprobs {} are nan".format(
                    len(lnp), np.sum(np.isnan(lnp))))
        if any(np.isposinf(lnp)):
            logging.warning(
                "Of {} lnprobs {} are +np.inf".format(
                    len(lnp), np.sum(np.isposinf(lnp))))
        if any(np.isneginf(lnp)):
            logging.warning(
                "Of {} lnprobs {} are -np.inf".format(
                    len(lnp), np.sum(np.isneginf(lnp))))
983

984
985
986
        lnp_finite = copy.copy(lnp)
        lnp_finite[np.isinf(lnp)] = np.nan
        p = pF[np.nanargmax(lnp_finite)]
987
988
        logging.info('Generating new p0 from max lnp which had twoF={}'
                     .format(lnl[np.nanargmax(lnp_finite)]))
989
        p0 = self.generate_scattered_p0(p)
990
991
992
993
994
995

        return p0

    def get_save_data_dictionary(self):
        d = dict(nsteps=self.nsteps, nwalkers=self.nwalkers,
                 ntemps=self.ntemps, theta_keys=self.theta_keys,
Gregory Ashton's avatar
Gregory Ashton committed
996
                 theta_prior=self.theta_prior, scatter_val=self.scatter_val,
997
998
                 log10temperature_min=self.log10temperature_min,
                 theta0_idx=self.theta0_idx)
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
        return d

    def save_data(self, sampler, samples, lnprobs, lnlikes):
        d = self.get_save_data_dictionary()
        d['sampler'] = sampler
        d['samples'] = samples
        d['lnprobs'] = lnprobs
        d['lnlikes'] = lnlikes

        if os.path.isfile(self.pickle_path):
            logging.info('Saving backup of {} as {}.old'.format(
                self.pickle_path, self.pickle_path))
            os.rename(self.pickle_path, self.pickle_path+".old")
        with open(self.pickle_path, "wb") as File:
            pickle.dump(d, File)

    def get_list_of_matching_sfts(self):
1016
        matches = glob.glob(self.sftfilepath)
1017
1018
1019
1020
        if len(matches) > 0:
            return matches
        else:
            raise IOError('No sfts found matching {}'.format(
1021
                self.sftfilepath))
1022
1023
1024
1025
1026
1027
1028

    def get_saved_data(self):
        with open(self.pickle_path, "r") as File:
            d = pickle.load(File)
        return d

    def check_old_data_is_okay_to_use(self):
1029
1030
1031
1032
        if args.use_old_data:
            logging.info("Forcing use of old data")
            return True

1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
        if os.path.isfile(self.pickle_path) is False:
            logging.info('No pickled data found')
            return False

        oldest_sft = min([os.path.getmtime(f) for f in
                          self.get_list_of_matching_sfts()])
        if os.path.getmtime(self.pickle_path) < oldest_sft:
            logging.info('Pickled data outdates sft files')
            return False

        old_d = self.get_saved_data().copy()
        new_d = self.get_save_data_dictionary().copy()

        old_d.pop('samples')
        old_d.pop('sampler')
        old_d.pop('lnprobs')
        old_d.pop('lnlikes')

        mod_keys = []
        for key in new_d.keys():
            if key in old_d:
                if new_d[key] != old_d[key]:
                    mod_keys.append((key, old_d[key], new_d[key]))
            else:
1057
                raise ValueError('Keys {} not in old dictionary'.format(key))
1058
1059
1060
1061
1062
1063
1064
1065
1066

        if len(mod_keys) == 0:
            return True
        else:
            logging.warning("Saved data differs from requested")
            logging.info("Differences found in following keys:")
            for key in mod_keys:
                if len(key) == 3:
                    if np.isscalar(key[1]) or key[0] == 'nsteps':
1067
                        logging.info("    {} : {} -> {}".format(*key))
1068
                    else:
1069
                        logging.info("    " + key[0])
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
                else:
                    logging.info(key)
            return False

    def get_max_twoF(self, threshold=0.05):
        """ Returns the max 2F sample and the corresponding 2F value

        Note: the sample is returned as a dictionary along with an estimate of
        the standard deviation calculated from the std of all samples with a
        twoF within `threshold` (relative) to the max twoF

        """
        if any(np.isposinf(self.lnlikes)):
            logging.info('twoF values contain positive infinite values')
        if any(np.isneginf(self.lnlikes)):
            logging.info('twoF values contain negative infinite values')
        if any(np.isnan(self.lnlikes)):
            logging.info('twoF values contain nan')
        idxs = np.isfinite(self.lnlikes)
        jmax = np.nanargmax(self.lnlikes[idxs])
        maxtwoF = self.lnlikes[jmax]
1091
        d = OrderedDict()
1092

Gregory Ashton's avatar
Gregory Ashton committed
1093
        repeats = []
1094
        for i, k in enumerate(self.theta_keys):
Gregory Ashton's avatar
Gregory Ashton committed
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
            if k in d and k not in repeats:
                d[k+'_0'] = d[k]  # relabel the old key
                d.pop(k)
                repeats.append(k)
            if k in repeats:
                k = k + '_0'
                count = 1
                while k in d:
                    k = k.replace('_{}'.format(count-1), '_{}'.format(count))
                    count += 1
1105
1106
1107
1108
1109
            d[k] = self.samples[jmax][i]
        return d, maxtwoF

    def get_median_stds(self):
        """ Returns a dict of the median and std of all production samples """
1110
        d = OrderedDict()
Gregory Ashton's avatar
Gregory Ashton committed
1111
        repeats = []
1112
        for s, k in zip(self.samples.T, self.theta_keys):
Gregory Ashton's avatar
Gregory Ashton committed
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
            if k in d and k not in repeats:
                d[k+'_0'] = d[k]  # relabel the old key
                d[k+'_0_std'] = d[k+'_std']
                d.pop(k)
                d.pop(k+'_std')
                repeats.append(k)
            if k in repeats:
                k = k + '_0'
                count = 1
                while k in d:
                    k = k.replace('_{}'.format(count-1), '_{}'.format(count))
                    count += 1

1126
1127
1128
1129
1130
1131
1132
1133
            d[k] = np.median(s)
            d[k+'_std'] = np.std(s)
        return d

    def write_par(self, method='med'):
        """ Writes a .par of the best-fit params with an estimated std """
        logging.info('Writing {}/{}.par using the {} method'.format(
            self.outdir, self.label, method))
1134
1135
1136
1137
1138
1139
1140

        median_std_d = self.get_median_stds()
        max_twoF_d, max_twoF = self.get_max_twoF()

        filename = '{}/{}.par'.format(self.outdir, self.label)
        with open(filename, 'w+') as f:
            f.write('MaxtwoF = {}\n'.format(max_twoF))
1141
            f.write('theta0_index = {}\n'.format(self.theta0_idx))
1142
            if method == 'med':
1143
1144
                for key, val in median_std_d.iteritems():
                    f.write('{} = {:1.16e}\n'.format(key, val))
1145
            if method == 'twoFmax':
1146
1147
1148
1149
                for key, val in max_twoF_d.iteritems():
                    f.write('{} = {:1.16e}\n'.format(key, val))

    def print_summary(self):
Gregory Ashton's avatar
Gregory Ashton committed
1150
        max_twoFd, max_twoF = self.get_max_twoF()
1151
        median_std_d = self.get_median_stds()
Gregory Ashton's avatar
Gregory Ashton committed
1152
        print('\nSummary:')
1153
        print('theta0 index: {}'.format(self.theta0_idx))
Gregory Ashton's avatar
Gregory Ashton committed
1154
1155
1156
1157
        print('Max twoF: {} with parameters:'.format(max_twoF))
        for k in np.sort(max_twoFd.keys()):
            print('  {:10s} = {:1.9e}'.format(k, max_twoFd[k]))
        print('\nMedian +/- std for production values')
1158
        for k in np.sort(median_std_d.keys()):
1159
            if 'std' not in k:
Gregory Ashton's avatar
Gregory Ashton committed
1160
                print('  {:10s} = {:1.9e} +/- {:1.9e}'.format(
1161
                    k, median_std_d[k], median_std_d[k+'_std']))
1162
1163


Gregory Ashton's avatar
Gregory Ashton committed
1164
1165
1166
class MCMCGlitchSearch(MCMCSearch):
    """ MCMC search using the SemiCoherentGlitchSearch """
    @initializer
1167
    def __init__(self, label, outdir, sftfilepath, theta_prior, tref,
1168
1169
                 tstart, tend, nglitch=1, nsteps=[100, 100, 100], nwalkers=100,
                 ntemps=1, log10temperature_min=-5, theta_initial=None,
1170
                 scatter_val=1e-4, dtglitchmin=1*86400, theta0_idx=0,
1171
                 detector=None, BSGL=False,
1172
                 minCoverFreq=None, maxCoverFreq=None, earth_ephem=None,
Gregory Ashton's avatar
Gregory Ashton committed
1173
1174
1175
1176
1177
                 sun_ephem=None):
        """
        Parameters
        label, outdir: str
            A label and directory to read/write data from/to
1178
1179
_        sftfilepath: str
            File patern to match SFTs
Gregory Ashton's avatar
Gregory Ashton committed
1180
1181
1182
1183
1184
1185
1186
1187
1188
        theta_prior: dict
            Dictionary of priors and fixed values for the search parameters.
            For each parameters (key of the dict), if it is to be held fixed
            the value should be the constant float, if it is be searched, the
            value should be a dictionary of the prior.
        theta_initial: dict, array, (None)
            Either a dictionary of distribution about which to distribute the
            initial walkers about, an array (from which the walkers will be
            scattered by scatter_val, or  None in which case the prior is used.
1189
1190
1191
1192
        scatter_val, float or ndim array
            Size of scatter to use about the initialisation step, if given as
            an array it must be of length ndim and the order is given by
            theta_keys
Gregory Ashton's avatar
Gregory Ashton committed
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
        nglitch: int
            The number of glitches to allow
        tref, tstart, tend: int
            GPS seconds of the reference time, start time and end time
        nsteps: list (m,)
            List specifying the number of steps to take, the last two entries
            give the nburn and nprod of the 'production' run, all entries
            before are for iterative initialisation steps (usually just one)
            e.g. [1000, 1000, 500].
        dtglitchmin: int
            The minimum duration (in seconds) of a segment between two glitches
            or a glitch and the start/end of the data
1205
1206
1207
1208
1209
1210
        nwalkers, ntemps: int,
            The number of walkers and temperates to use in the parallel
            tempered PTSampler.
        log10temperature_min float < 0
            The  log_10(tmin) value, the set of betas passed to PTSampler are
            generated from np.logspace(0, log10temperature_min, ntemps).
1211
1212
1213
1214
        theta0_idx, int
            Index (zero-based) of which segment the theta refers to - uyseful
            if providing a tight prior on theta to allow the signal to jump
            too theta (and not just from)
1215
1216
1217
        detector: str
            Two character reference to the data to use, specify None for no
            contraint.
Gregory Ashton's avatar
Gregory Ashton committed
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
        minCoverFreq, maxCoverFreq: float
            Minimum and maximum instantaneous frequency which will be covered
            over the SFT time span as passed to CreateFstatInput
        earth_ephem, sun_ephem: str
            Paths of the two files containing positions of Earth and Sun,
            respectively at evenly spaced times, as passed to CreateFstatInput
            If None defaults defined in BaseSearchClass will be used

        """

1228
        self.add_log_file()
Gregory Ashton's avatar
Gregory Ashton committed
1229
1230
        logging.info(('Set-up MCMC glitch search with {} glitches for model {}'
                      ' on data {}').format(self.nglitch, self.label,
1231
                                            self.sftfilepath))
Gregory Ashton's avatar
Gregory Ashton committed
1232
1233
        if os.path.isdir(outdir) is False:
            os.mkdir(outdir)
1234
1235
        self.minStartTime = tstart
        self.maxStartTime = tend
Gregory Ashton's avatar
Gregory Ashton committed
1236
1237
1238
        self.pickle_path = '{}/{}_saved_data.p'.format(self.outdir, self.label)
        self.unpack_input_theta()
        self.ndim = len(self.theta_keys)
1239
1240
1241
1242
        if self.log10temperature_min:
            self.betas = np.logspace(0, self.log10temperature_min, self.ntemps)
        else:
            self.betas = None
Gregory Ashton's avatar
Gregory Ashton committed
1243
1244
1245
1246
1247
1248
1249
1250
1251
        if earth_ephem is None:
            self.earth_ephem = self.earth_ephem_default
        if sun_ephem is None:
            self.sun_ephem = self.sun_ephem_default

        if args.clean and os.path.isfile(self.pickle_path):
            os.rename(self.pickle_path, self.pickle_path+".old")

        self.old_data_is_okay_to_use = self.check_old_data_is_okay_to_use()
1252
        self.log_input()
Gregory Ashton's avatar
Gregory Ashton committed
1253
1254
1255
1256

    def inititate_search_object(self):
        logging.info('Setting up search object')
        self.search = SemiCoherentGlitchSearch(
1257
1258
            label=self.label, outdir=self.outdir, sftfilepath=self.sftfilepath,
            tref=self.tref, tstart=self.tstart,
Gregory Ashton's avatar
Gregory Ashton committed
1259
1260
            tend=self.tend, minCoverFreq=self.minCoverFreq,
            maxCoverFreq=self.maxCoverFreq, earth_ephem=self.earth_ephem,
1261
            sun_ephem=self.sun_ephem, detector=self.detector, BSGL=self.BSGL,
1262
1263
            nglitch=self.nglitch, theta0_idx=self.theta0_idx,
            minStartTime=self.minStartTime, maxStartTime=self.maxStartTime)
Gregory Ashton's avatar
Gregory Ashton committed
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345

    def logp(self, theta_vals, theta_prior, theta_keys, search):
        if self.nglitch > 1:
            ts = [self.tstart] + theta_vals[-self.nglitch:] + [self.tend]
            if np.array_equal(ts, np.sort(ts)) is False:
                return -np.inf
            if any(np.diff(ts) < self.dtglitchmin):
                return -np.inf

        H = [self.generic_lnprior(**theta_prior[key])(p) for p, key in
             zip(theta_vals, theta_keys)]
        return np.sum(H)

    def logl(self, theta, search):
        for j, theta_i in enumerate(self.theta_idxs):
            self.fixed_theta[theta_i] = theta[j]
        FS = search.compute_nglitch_fstat(*self.fixed_theta)
        return FS

    def unpack_input_theta(self):
        glitch_keys = ['delta_F0', 'delta_F1', 'tglitch']
        full_glitch_keys = list(np.array(
            [[gk]*self.nglitch for gk in glitch_keys]).flatten())
        full_theta_keys = ['F0', 'F1', 'F2', 'Alpha', 'Delta']+full_glitch_keys
        full_theta_keys_copy = copy.copy(full_theta_keys)

        glitch_symbols = ['$\delta f$', '$\delta \dot{f}$', r'$t_{glitch}$']
        full_glitch_symbols = list(np.array(
            [[gs]*self.nglitch for gs in glitch_symbols]).flatten())
        full_theta_symbols = (['$f$', '$\dot{f}$', '$\ddot{f}$', r'$\alpha$',
                               r'$\delta$'] + full_glitch_symbols)
        self.theta_keys = []
        fixed_theta_dict = {}
        for key, val in self.theta_prior.iteritems():
            if type(val) is dict:
                fixed_theta_dict[key] = 0
                if key in glitch_keys:
                    for i in range(self.nglitch):
                        self.theta_keys.append(key)
                else:
                    self.theta_keys.append(key)
            elif type(val) in [float, int, np.float64]:
                fixed_theta_dict[key] = val
            else:
                raise ValueError(
                    'Type {} of {} in theta not recognised'.format(
                        type(val), key))
            if key in glitch_keys:
                for i in range(self.nglitch):
                    full_theta_keys_copy.pop(full_theta_keys_copy.index(key))
            else:
                full_theta_keys_copy.pop(full_theta_keys_copy.index(key))

        if len(full_theta_keys_copy) > 0:
            raise ValueError(('Input dictionary `theta` is missing the'
                              'following keys: {}').format(
                                  full_theta_keys_copy))

        self.fixed_theta = [fixed_theta_dict[key] for key in full_theta_keys]
        self.theta_idxs = [full_theta_keys.index(k) for k in self.theta_keys]
        self.theta_symbols = [full_theta_symbols[i] for i in self.theta_idxs]

        idxs = np.argsort(self.theta_idxs)
        self.theta_idxs = [self.theta_idxs[i] for i in idxs]
        self.theta_symbols = [self.theta_symbols[i] for i in idxs]
        self.theta_keys = [self.theta_keys[i] for i in idxs]

        # Correct for number of glitches in the idxs
        self.theta_idxs = np.array(self.theta_idxs)
        while np.sum(self.theta_idxs[:-1] == self.theta_idxs[1:]) > 0:
            for i, idx in enumerate(self.theta_idxs):
                if idx in self.theta_idxs[:i]:
                    self.theta_idxs[i] += 1

    def apply_corrections_to_p0(self, p0):
        p0 = np.array(p0)
        if self.nglitch > 1:
            p0[:, :, -self.nglitch:] = np.sort(p0[:, :, -self.nglitch:],
                                               axis=2)
        return p0


Gregory Ashton's avatar
Gregory Ashton committed
1346
1347
class GridSearch(BaseSearchClass):
    """ Gridded search using ComputeFstat """
1348
    @initializer
1349
    def __init__(self, label, outdir, sftfilepath, F0s=[0],
Gregory Ashton's avatar
Gregory Ashton committed
1350
1351
                 F1s=[0], F2s=[0], Alphas=[0], Deltas=[0], tref=None,
                 tstart=None, tend=None, minCoverFreq=None, maxCoverFreq=None,
1352
                 earth_ephem=None, sun_ephem=None, detector=None, BSGL=False):
1353
1354
1355
1356
        """
        Parameters
        label, outdir: str
            A label and directory to read/write data from/to
1357
1358
        sftfilepath: str
            File patern to match SFTs
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
        F0s, F1s, F2s, delta_F0s, delta_F1s, tglitchs, Alphas, Deltas: tuple
            Length 3 tuple describing the grid for each parameter, e.g
            [F0min, F0max, dF0], for a fixed value simply give [F0].
        tref, tstart, tend: int
            GPS seconds of the reference time, start time and end time
        minCoverFreq, maxCoverFreq: float
            Minimum and maximum instantaneous frequency which will be covered
            over the SFT time span as passed to CreateFstatInput
        earth_ephem, sun_ephem: str
            Paths of the two files containing positions of Earth and Sun,
            respectively at evenly spaced times, as passed to CreateFstatInput