pyfstat.py 61.9 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
16
17
18
19
20
21
22

import numpy as np
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import emcee
import corner
import dill as pickle
import lalpulsar

23
24
plt.rcParams['text.usetex'] = True

25
26
27
28
29
30
31
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(' ', '')
32
            v = v.replace(' ', '').replace("'", "").replace('"', '').replace('\n', '')
33
34
35
36
37
38
            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')
39
40
41
    earth_ephem = None
    sun_ephem = None

42
43
44
45
46
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")
47
parser.add_argument("-u", "--use-old-data", action="store_true")
48
49
50
51
52
53
54
55
56
57
58
59
60
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')

61
62

def initializer(func):
63
    """ Automatically assigns the parameters to self """
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
    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):
81
    """ Read in a .par file, returns a dictionary of the values """
82
83
84
85
86
87
88
89
90
91
    filename = '{}/{}.par'.format(outdir, label)
    d = {}
    with open(filename, 'r') as f:
        for line in f:
            key, val = line.rstrip('\n').split(' = ')
            d[key] = np.float64(val)
    return d


class BaseSearchClass(object):
92
    """ The base search class, provides ephemeris and general utilities """
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

    earth_ephem_default = earth_ephem
    sun_ephem_default = sun_ephem

    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
122
            lowest degree e.g [phi, F0, F1,...].
123
        dT: float
124
            difference between the two reference times as tref_new - tref_old.
125
126
127
128

        Returns
        -------
        theta_new: array-like shape (n,)
129
            vector of the coefficients as evaluate as the new reference time.
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
        """
        n = len(theta)
        m = self.shift_matrix(n, dT)
        return np.dot(m, theta)

    def calculate_thetas(self, theta, delta_thetas, tbounds):
        """ Calculates the set of coefficients for the post-glitch signal """
        thetas = [theta]
        for i, dt in enumerate(delta_thetas):
            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]))
        return thetas


Gregory Ashton's avatar
Gregory Ashton committed
147
148
149
150
151
152
153
154
155
class ComputeFstat(object):
    """ Base class providing interface to lalpulsar.ComputeFstat """

    earth_ephem_default = earth_ephem
    sun_ephem_default = sun_ephem

    @initializer
    def __init__(self, tref, sftlabel=None, sftdir=None,
                 minCoverFreq=None, maxCoverFreq=None,
156
157
                 detector=None, earth_ephem=None, sun_ephem=None,
                 binary=False, transient=True):
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
        """
        Parameters
        ----------
        tref: int
            GPS seconds of the reference time.
        sftlabel, sftdir: str
            A label and directory in which to find the relevant sft file
        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.

        """
Gregory Ashton's avatar
Gregory Ashton committed
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199

        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
        self.sft_filepath = self.sftdir+'/*_'+self.sftlabel+"*sft"
        SFTCatalog = lalpulsar.SFTdataFind(self.sft_filepath, constraints)
        names = list(set([d.header.name for d in SFTCatalog.data]))
200
201
202
        logging.info(
            'Loaded data from detectors {} matching pattern {}'.format(
                names, self.sft_filepath))
Gregory Ashton's avatar
Gregory Ashton committed
203
204
205
206
207
208

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

        logging.info('Initialising FstatInput')
        dFreq = 0
209
210
211
212
213
        if self.transient:
            self.whatToCompute = lalpulsar.FSTATQ_ATOMS_PER_DET
        else:
            self.whatToCompute = lalpulsar.FSTATQ_2F

Gregory Ashton's avatar
Gregory Ashton committed
214
215
216
217
218
219
220
221
        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
222
223
224
            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
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244

        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()

245
246
247
248
249
250
251
        if self.transient:
            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
252

Gregory Ashton's avatar
Gregory Ashton committed
253
    def run_computefstatistic_single_point(self, tstart, tend, F0, F1,
254
255
256
                                           F2, Alpha, Delta, asini=None,
                                           period=None, ecc=None, tp=None,
                                           argp=None):
257
        """ Returns the twoF fully-coherently at a single point """
Gregory Ashton's avatar
Gregory Ashton committed
258
259
260
261

        self.PulsarDopplerParams.fkdot = np.array([F0, F1, F2, 0, 0, 0, 0])
        self.PulsarDopplerParams.Alpha = Alpha
        self.PulsarDopplerParams.Delta = Delta
262
263
264
265
266
267
        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
268
269
270
271

        lalpulsar.ComputeFstat(self.FstatResults,
                               self.FstatInput,
                               self.PulsarDopplerParams,
272
                               1,
Gregory Ashton's avatar
Gregory Ashton committed
273
274
275
                               self.whatToCompute
                               )

276
277
278
        if self.transient is False:
            return self.FstatResults.twoF[0]

279
280
        self.windowRange.t0 = int(tstart)  # TYPE UINT4
        self.windowRange.tau = int(tend - tstart)  # TYPE UINT4
Gregory Ashton's avatar
Gregory Ashton committed
281
        FS = lalpulsar.ComputeTransientFstatMap(
282
            self.FstatResults.multiFatoms[0], self.windowRange, False)
Gregory Ashton's avatar
Gregory Ashton committed
283
284
285
286
        return 2*FS.F_mn.data[0][0]


class SemiCoherentGlitchSearch(BaseSearchClass, ComputeFstat):
287
288
289
290
291
292
293
294
295
    """ 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
296
    def __init__(self, label, outdir, tref, tstart, tend, nglitch=0,
297
298
299
                 sftlabel=None, sftdir=None, minCoverFreq=None,
                 maxCoverFreq=None, detector=None, earth_ephem=None,
                 sun_ephem=None):
300
301
302
303
        """
        Parameters
        ----------
        label, outdir: str
304
305
306
307
308
309
            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).
310
311
        sftlabel, sftdir: str
            A label and directory in which to find the relevant sft file. If
312
            None use label and outdir.
313
        minCoverFreq, maxCoverFreq: float
314
315
316
            The min and max cover frequency passed to CreateFstatInput, if
            either is None the range of frequencies in the SFT less 1Hz is
            used.
317
318
        detector: str
            Two character reference to the data to use, specify None for no
319
            contraint.
320
321
        earth_ephem, sun_ephem: str
            Paths of the two files containing positions of Earth and Sun,
322
323
            respectively at evenly spaced times, as passed to CreateFstatInput.
            If None defaults defined in BaseSearchClass will be used.
324
325
326
327
328
329
330
331
332
333
334
        """

        if self.sftlabel is None:
            self.sftlabel = self.label
        if self.sftdir is None:
            self.sftdir = self.outdir
        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
335
336
        self.transient = True
        self.binary = False
337
338
339
        self.init_computefstatistic_single_point()

    def compute_nglitch_fstat(self, F0, F1, F2, Alpha, Delta, *args):
340
        """ Returns the semi-coherent glitch summed twoF """
341
342
343

        args = list(args)
        tboundaries = [self.tstart] + args[-self.nglitch:] + [self.tend]
344
345
346
347
348
349
350
351
352
        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)

        thetas = self.calculate_thetas(theta, delta_thetas, tboundaries)
353
354

        twoFSum = 0
355
        for i, theta_i_at_tref in enumerate(thetas):
356
357
358
            ts, te = tboundaries[i], tboundaries[i+1]

            twoFVal = self.run_computefstatistic_single_point(
359
360
                ts, te, theta_i_at_tref[1], theta_i_at_tref[2],
                theta_i_at_tref[3], Alpha, Delta)
361
362
363
364
365
366
            twoFSum += twoFVal

        return twoFSum

    def compute_glitch_fstat_single(self, F0, F1, F2, Alpha, Delta, delta_F0,
                                    delta_F1, tglitch):
367
368
369
370
        """ Returns the semi-coherent glitch summed twoF for nglitch=1

        Note: used for testing
        """
371
372
373
374
375
376
377
378
379
380
381

        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
382
            self.tstart, tglitch, theta[0], theta[1], theta[2], Alpha,
383
384
385
386
387
388
            Delta)

        if tglitch == self.tend:
            return twoFsegA

        twoFsegB = self.run_computefstatistic_single_point(
Gregory Ashton's avatar
Gregory Ashton committed
389
            tglitch, self.tend, theta_post_glitch[0],
390
391
392
393
394
395
            theta_post_glitch[1], theta_post_glitch[2], Alpha,
            Delta)

        return twoFsegA + twoFsegB


Gregory Ashton's avatar
Gregory Ashton committed
396
397
class MCMCSearch(BaseSearchClass):
    """ MCMC search using ComputeFstat"""
398
    @initializer
399
400
    def __init__(self, label, outdir, sftlabel, sftdir, theta_prior, tref,
                 tstart, tend, nsteps=[100, 100, 100], nwalkers=100, ntemps=1,
401
402
                 log10temperature_min=-5, theta_initial=None, scatter_val=1e-4,
                 binary=False, minCoverFreq=None, maxCoverFreq=None,
Gregory Ashton's avatar
Gregory Ashton committed
403
                 detector=None, earth_ephem=None, sun_ephem=None):
404
405
406
407
408
409
        """
        Parameters
        label, outdir: str
            A label and directory to read/write data from/to
        sftlabel, sftdir: str
            A label and directory in which to find the relevant sft file
410
        theta_prior: dict
411
412
413
414
            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.
415
416
417
418
        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.
419
420
421
422
423
424
425
        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].
426
427
428
429
430
431
432
433
434
435
436
        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.
437
438
439
440
441
442
443
444
445
446
        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

        """

Gregory Ashton's avatar
Gregory Ashton committed
447
448
449
        logging.info(
            'Set-up MCMC search for model {} on data {}'.format(
                self.label, self.sftlabel))
450
451
452
        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
453
454
        self.theta_prior['tstart'] = self.tstart
        self.theta_prior['tend'] = self.tend
455
456
        self.unpack_input_theta()
        self.ndim = len(self.theta_keys)
457
        self.betas = np.logspace(0, self.log10temperature_min, self.ntemps)
458
        self.sft_filepath = self.sftdir+'/*_'+self.sftlabel+"*sft"
459

460
461
462
463
464
465
466
467
468
469
470
471
        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()

    def inititate_search_object(self):
        logging.info('Setting up search object')
Gregory Ashton's avatar
Gregory Ashton committed
472
473
474
        self.search = ComputeFstat(
            tref=self.tref, sftlabel=self.sftlabel,
            sftdir=self.sftdir, minCoverFreq=self.minCoverFreq,
475
            maxCoverFreq=self.maxCoverFreq, earth_ephem=self.earth_ephem,
476
            sun_ephem=self.sun_ephem, detector=self.detector, transient=False)
477
478

    def logp(self, theta_vals, theta_prior, theta_keys, search):
Gregory Ashton's avatar
Gregory Ashton committed
479
        H = [self.generic_lnprior(**theta_prior[key])(p) for p, key in
480
481
482
483
484
485
             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
486
        FS = search.run_computefstatistic_single_point(*self.fixed_theta)
487
488
489
        return FS

    def unpack_input_theta(self):
Gregory Ashton's avatar
Gregory Ashton committed
490
491
        full_theta_keys = ['tstart', 'tend', 'F0', 'F1', 'F2', 'Alpha',
                           'Delta']
492
493
494
        if self.binary:
            full_theta_keys += [
                'asini', 'period', 'ecc', 'tp', 'argp']
495
496
        full_theta_keys_copy = copy.copy(full_theta_keys)

Gregory Ashton's avatar
Gregory Ashton committed
497
498
        full_theta_symbols = ['_', '_', '$f$', '$\dot{f}$', '$\ddot{f}$',
                              r'$\alpha$', r'$\delta$']
499
500
501
502
        if self.binary:
            full_theta_symbols += [
                'asini', 'period', 'period', 'ecc', 'tp', 'argp']

503
504
        self.theta_keys = []
        fixed_theta_dict = {}
505
        for key, val in self.theta_prior.iteritems():
506
507
            if type(val) is dict:
                fixed_theta_dict[key] = 0
Gregory Ashton's avatar
Gregory Ashton committed
508
                self.theta_keys.append(key)
509
510
511
512
513
514
            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
515
            full_theta_keys_copy.pop(full_theta_keys_copy.index(key))
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559

        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):
        initial_priors = np.array([
            self.logp(p, self.theta_prior, self.theta_keys, self.search)
            for p in p0[0]])
        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))

    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
560
561
        p0 = self.generate_initial_p0()
        p0 = self.apply_corrections_to_p0(p0)
562
563
564
565
566
567
568
        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(
                j, ninit_steps, n))
            sampler.run_mcmc(p0, n)
569
570
            logging.info("Mean acceptance fraction: {0:.3f}"
                         .format(np.mean(sampler.acceptance_fraction)))
Gregory Ashton's avatar
Gregory Ashton committed
571
            fig, axes = self.plot_walkers(sampler, symbols=self.theta_symbols)
572
573
574
            fig.savefig('{}/{}_init_{}_walkers.png'.format(
                self.outdir, self.label, j))

575
            p0 = self.get_new_p0(sampler)
Gregory Ashton's avatar
Gregory Ashton committed
576
            p0 = self.apply_corrections_to_p0(p0)
577
578
579
580
581
582
583
584
            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)
585
586
        logging.info("Mean acceptance fraction: {0:.3f}"
                     .format(np.mean(sampler.acceptance_fraction)))
587

Gregory Ashton's avatar
Gregory Ashton committed
588
        fig, axes = self.plot_walkers(sampler, symbols=self.theta_symbols)
589
590
591
592
593
594
595
596
597
598
599
        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)

600
    def plot_corner(self, figsize=(7, 7),  tglitch_ratio=False,
601
602
603
                    add_prior=False, nstds=None, label_offset=0.4, **kwargs):

        fig, axes = plt.subplots(self.ndim, self.ndim,
604
                                 figsize=figsize)
605
606
607
608
609
610

        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]

611
612
613
614
615
616
617
        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'$R_{\textrm{glitch}}$'
618
619
620
621
622
623
624
625
626
627
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
664

        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))

    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
665
            prior = self.generic_lnprior(**self.theta_prior[key])
666
667
668
669
670
671
            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
672
    def generic_lnprior(self, **kwargs):
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
        """ 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'])
        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
722
    def generate_rv(self, **kwargs):
723
724
725
726
727
728
729
730
731
732
733
734
735
736
        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']))
        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
737
738
    def plot_walkers(self, sampler, symbols=None, alpha=0.4, color="k", temp=0,
                     start=None, stop=None, draw_vline=None):
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
        """ 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):
759
                    axes[i].ticklabel_format(useOffset=False, axis='y')
760
761
                    cs = chain[:, start:stop, i].T
                    axes[i].plot(cs, color="k", alpha=alpha)
762
763
764
765
                    if symbols:
                        axes[i].set_ylabel(symbols[i])
                    if draw_vline is not None:
                        axes[i].axvline(draw_vline, lw=2, ls="--")
766
767
768
769
770

            else:
                cs = chain[:, start:stop, 0].T
                axes.plot(cs, color='k', alpha=alpha)
                axes.ticklabel_format(useOffset=False, axis='y')
771
772
773

        return fig, axes

Gregory Ashton's avatar
Gregory Ashton committed
774
775
776
777
778
    def apply_corrections_to_p0(self, p0):
        """ Apply any correction to the initial p0 values """
        return p0

    def generate_scattered_p0(self, p):
779
        """ Generate a set of p0s scattered about p """
Gregory Ashton's avatar
Gregory Ashton committed
780
        p0 = [[p + self.scatter_val * p * np.random.randn(self.ndim)
781
782
783
784
               for i in xrange(self.nwalkers)]
              for j in xrange(self.ntemps)]
        return p0

Gregory Ashton's avatar
Gregory Ashton committed
785
    def generate_initial_p0(self):
786
787
788
        """ Generate a set of init vals for the walkers """

        if type(self.theta_initial) == dict:
Gregory Ashton's avatar
Gregory Ashton committed
789
            p0 = [[[self.generate_rv(**self.theta_initial[key])
790
791
792
793
                    for key in self.theta_keys]
                   for i in range(self.nwalkers)]
                  for j in range(self.ntemps)]
        elif self.theta_initial is None:
Gregory Ashton's avatar
Gregory Ashton committed
794
            p0 = [[[self.generate_rv(**self.theta_prior[key])
795
796
797
798
                    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
799
            p0 = self.generate_scattered_p0(self.theta_initial)
800
801
802
803
804
        else:
            raise ValueError('theta_initial not understood')

        return p0

805
    def get_new_p0(self, sampler):
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
        """ 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, :, :]
        lnp = sampler.lnprobability[:, :, -1].reshape(
            self.ntemps, self.nwalkers)[0, :]
        if any(np.isnan(lnp)):
            logging.warning("The sampler has produced nan's")

        p = pF[np.nanargmax(lnp)]
824
        p0 = self.generate_scattered_p0(p)
825
826
827
828
829
830

        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
831
832
                 theta_prior=self.theta_prior, scatter_val=self.scatter_val,
                 log10temperature_min=self.log10temperature_min)
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
        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):
        matches = glob.glob(self.sft_filepath)
        if len(matches) > 0:
            return matches
        else:
            raise IOError('No sfts found matching {}'.format(
                self.sft_filepath))

    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):
863
864
865
866
        if args.use_old_data:
            logging.info("Forcing use of old data")
            return True

867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
        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:
                raise ValueError('Keys do not match')

        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':
901
                        logging.info("    {} : {} -> {}".format(*key))
902
                    else:
903
                        logging.info("    " + key[0])
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
                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]
925
        d = OrderedDict()
926
927
928
929
930
931
932
933
934
935
936
937
938
939
        close_idxs = abs((maxtwoF - self.lnlikes[idxs]) / maxtwoF) < threshold
        for i, k in enumerate(self.theta_keys):
            base_key = copy.copy(k)
            ng = 1
            while k in d:
                k = base_key + '_{}'.format(ng)
            d[k] = self.samples[jmax][i]

            s = self.samples[:, i][close_idxs]
            d[k + '_std'] = np.std(s)
        return d, maxtwoF

    def get_median_stds(self):
        """ Returns a dict of the median and std of all production samples """
940
        d = OrderedDict()
941
942
943
944
945
946
947
948
949
        for s, k in zip(self.samples.T, self.theta_keys):
            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))
950
951
952
953
954
955
956
957

        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))
            if method == 'med':
958
959
                for key, val in median_std_d.iteritems():
                    f.write('{} = {:1.16e}\n'.format(key, val))
960
            if method == 'twoFmax':
961
962
963
964
965
966
967
968
969
970
971
972
                for key, val in max_twoF_d.iteritems():
                    f.write('{} = {:1.16e}\n'.format(key, val))

    def print_summary(self):
        d, max_twoF = self.get_max_twoF()
        print('Max twoF: {}'.format(max_twoF))
        for k in np.sort(d.keys()):
            if 'std' not in k:
                print('{:10s} = {:1.9e} +/- {:1.9e}'.format(
                    k, d[k], d[k+'_std']))


Gregory Ashton's avatar
Gregory Ashton committed
973
974
975
976
class MCMCGlitchSearch(MCMCSearch):
    """ MCMC search using the SemiCoherentGlitchSearch """
    @initializer
    def __init__(self, label, outdir, sftlabel, sftdir, theta_prior, tref,
977
978
                 tstart, tend, nglitch=1, nsteps=[100, 100, 100], nwalkers=100,
                 ntemps=1, log10temperature_min=-5, theta_initial=None,
979
                 scatter_val=1e-4, dtglitchmin=1*86400, detector=None,
980
                 minCoverFreq=None, maxCoverFreq=None, earth_ephem=None,
Gregory Ashton's avatar
Gregory Ashton committed
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
                 sun_ephem=None):
        """
        Parameters
        label, outdir: str
            A label and directory to read/write data from/to
        sftlabel, sftdir: str
            A label and directory in which to find the relevant sft file
        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.
997
998
999
1000
        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