mcmc_based_searches.py 91.2 KB
Newer Older
Gregory Ashton's avatar
Gregory Ashton committed
1
""" Searches using MCMC-based methods """
2
from __future__ import division, absolute_import, print_function
Gregory Ashton's avatar
Gregory Ashton committed
3

4
import sys
Gregory Ashton's avatar
Gregory Ashton committed
5
import os
6
import copy
Gregory Ashton's avatar
Gregory Ashton committed
7
import logging
8
from collections import OrderedDict
9
import subprocess
10
11
12
13
14
15
16
17

import numpy as np
import matplotlib
import matplotlib.pyplot as plt
import emcee
import corner
import dill as pickle

18
import pyfstat.core as core
19
20
from pyfstat.core import tqdm, args, read_par
import pyfstat.optimal_setup_functions as optimal_setup_functions
21
import pyfstat.helper_functions as helper_functions
22
23


24
class MCMCSearch(core.BaseSearchClass):
Gregory Ashton's avatar
Gregory Ashton committed
25
    """ MCMC search using ComputeFstat"""
26
27

    symbol_dictionary = dict(
28
        F0='$f$', F1='$\dot{f}$', F2='$\ddot{f}$', Alpha=r'$\alpha$',
29
30
        Delta='$\delta$', asini='asini', period='P', ecc='ecc', tp='tp',
        argp='argp')
31
    unit_dictionary = dict(
32
33
        F0='Hz', F1='Hz/s', F2='Hz/s$^2$', Alpha=r'rad', Delta='rad',
        asini='', period='s', ecc='', tp='', argp='')
34
35
    rescale_dictionary = {}

Gregory Ashton's avatar
Gregory Ashton committed
36
    @helper_functions.initializer
Gregory Ashton's avatar
Gregory Ashton committed
37
    def __init__(self, label, outdir, theta_prior, tref, minStartTime,
38
                 maxStartTime, sftfilepattern=None, nsteps=[100, 100],
39
                 nwalkers=100, ntemps=1, log10temperature_min=-5,
40
                 theta_initial=None, scatter_val=1e-10, rhohatmax=1000,
41
                 binary=False, BSGL=False, minCoverFreq=None, SSBprec=None,
42
43
                 maxCoverFreq=None, detectors=None,
                 injectSources=None, assumeSqrtSX=None):
44
45
        """
        Parameters
46
        ----------
47
48
        label, outdir: str
            A label and directory to read/write data from/to
49
        sftfilepattern: str
50
51
            Pattern to match SFTs using wildcards (*?) and ranges [0-9];
            mutiple patterns can be given separated by colons.
52
        theta_prior: dict
53
54
55
56
            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.
57
58
59
60
        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.
61
        tref, minStartTime, maxStartTime: int
62
63
64
65
66
67
            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].
68
69
70
71
72
73
        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).
74
75
76
77
        rhohatmax: float
            Upper bound for the SNR scale parameter (required to normalise the
            Bayes factor) - this needs to be carefully set when using the
            evidence.
78
79
        binary: Bool
            If true, search over binary parameters
80
        detectors: str
81
82
            Two character reference to the data to use, specify None for no
            contraint.
83
84
85
86
87
88
        minCoverFreq, maxCoverFreq: float
            Minimum and maximum instantaneous frequency which will be covered
            over the SFT time span as passed to CreateFstatInput

        """

Gregory Ashton's avatar
Gregory Ashton committed
89
90
        if os.path.isdir(outdir) is False:
            os.mkdir(outdir)
91
        self._add_log_file()
92
        logging.info('Set-up MCMC search for model {}'.format(self.label))
93
94
        if sftfilepattern:
            logging.info('Using data {}'.format(self.sftfilepattern))
95
        else:
96
            logging.info('No sftfilepattern given')
97
98
        if injectSources:
            logging.info('Inject sources: {}'.format(injectSources))
99
        self.pickle_path = '{}/{}_saved_data.p'.format(self.outdir, self.label)
100
        self._unpack_input_theta()
101
        self.ndim = len(self.theta_keys)
102
103
104
105
        if self.log10temperature_min:
            self.betas = np.logspace(0, self.log10temperature_min, self.ntemps)
        else:
            self.betas = None
106

107
108
109
        if args.clean and os.path.isfile(self.pickle_path):
            os.rename(self.pickle_path, self.pickle_path+".old")

110
111
112
113
        self._set_likelihoodcoef()

    def _set_likelihoodcoef(self):
        self.likelihoodcoef = np.log(70./self.rhohatmax**4)
114

115
        self._log_input()
116

117
    def _log_input(self):
118
        logging.info('theta_prior = {}'.format(self.theta_prior))
119
        logging.info('nwalkers={}'.format(self.nwalkers))
120
121
122
123
        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(
124
            self.log10temperature_min))
125

126
    def _initiate_search_object(self):
127
        logging.info('Setting up search object')
128
        self.search = core.ComputeFstat(
129
            tref=self.tref, sftfilepattern=self.sftfilepattern,
130
            minCoverFreq=self.minCoverFreq, maxCoverFreq=self.maxCoverFreq,
131
            detectors=self.detectors, BSGL=self.BSGL, transient=False,
132
            minStartTime=self.minStartTime, maxStartTime=self.maxStartTime,
133
            binary=self.binary, injectSources=self.injectSources,
134
            assumeSqrtSX=self.assumeSqrtSX, SSBprec=self.SSBprec)
135
136

    def logp(self, theta_vals, theta_prior, theta_keys, search):
137
        H = [self._generic_lnprior(**theta_prior[key])(p) for p, key in
138
139
140
141
142
143
             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]
144
145
        FS = search.compute_fullycoherent_det_stat_single_point(
            *self.fixed_theta)
146
        return FS + self.likelihoodcoef
147

148
    def _unpack_input_theta(self):
149
        full_theta_keys = ['F0', 'F1', 'F2', 'Alpha', 'Delta']
150
151
152
        if self.binary:
            full_theta_keys += [
                'asini', 'period', 'ecc', 'tp', 'argp']
153
154
        full_theta_keys_copy = copy.copy(full_theta_keys)

155
156
        full_theta_symbols = ['$f$', '$\dot{f}$', '$\ddot{f}$', r'$\alpha$',
                              r'$\delta$']
157
158
        if self.binary:
            full_theta_symbols += [
159
                'asini', 'period', 'ecc', 'tp', 'argp']
160

161
162
        self.theta_keys = []
        fixed_theta_dict = {}
163
        for key, val in self.theta_prior.iteritems():
164
165
            if type(val) is dict:
                fixed_theta_dict[key] = 0
Gregory Ashton's avatar
Gregory Ashton committed
166
                self.theta_keys.append(key)
167
168
169
170
171
172
            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
173
            full_theta_keys_copy.pop(full_theta_keys_copy.index(key))
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188

        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]

189
    def _check_initial_points(self, p0):
190
191
192
193
194
195
196
197
198
199
200
201
202
        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))

203
                p0 = self._generate_new_p0_to_fix_initial_points(
204
205
                    p0, nt, initial_priors)

206
    def _generate_new_p0_to_fix_initial_points(self, p0, nt, initial_priors):
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
        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
226

227
    def _OLD_run_sampler_with_progress_bar(self, sampler, ns, p0):
228
229
        for result in tqdm(sampler.sample(p0, iterations=ns), total=ns):
            pass
Gregory Ashton's avatar
Gregory Ashton committed
230
231
        return sampler

232
233
    def setup_burnin_convergence_testing(
            self, n=10, test_type='autocorr', windowed=False, **kwargs):
234
235
236
237
238
        """
        If called, convergence testing is used during the MCMC simulation

        Parameters
        ----------
239
240
241
242
243
244
245
246
247
        n: int
            Number of steps after which to test convergence
        test_type: str ['autocorr', 'GR']
            If 'autocorr' use the exponential autocorrelation time (kwargs
            passed to `get_autocorr_convergence`). If 'GR' use the Gelman-Rubin
            statistic (kwargs passed to `get_GR_convergence`)
        windowed: bool
            If True, only calculate the convergence test in a window of length
            `n`
248
        """
249
        logging.info('Setting up convergence testing')
250
251
252
253
        self.convergence_n = n
        self.convergence_windowed = windowed
        self.convergence_test_type = test_type
        self.convergence_kwargs = kwargs
254
255
        self.convergence_diagnostic = []
        self.convergence_diagnosticx = []
256
257
258
        if test_type in ['autocorr']:
            self._get_convergence_test = self.test_autocorr_convergence
        elif test_type in ['GR']:
Gregory Ashton's avatar
Gregory Ashton committed
259
            self._get_convergence_test = self.test_GR_convergence
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
        else:
            raise ValueError('test_type {} not understood'.format(test_type))

    def test_autocorr_convergence(self, i, sampler, test=True, n_cut=5):
        try:
            acors = np.zeros((self.ntemps, self.ndim))
            for temp in range(self.ntemps):
                if self.convergence_windowed:
                    j = i-self.convergence_n
                else:
                    j = 0
                x = np.mean(sampler.chain[temp, :, j:i, :], axis=0)
                acors[temp, :] = emcee.autocorr.exponential_time(x)
            c = np.max(acors, axis=0)
        except emcee.autocorr.AutocorrError:
Gregory Ashton's avatar
Gregory Ashton committed
275
276
277
278
            logging.info('Failed to calculate exponential autocorrelation')
            c = np.zeros(self.ndim) + np.nan
        except AttributeError:
            logging.info('Unable to calculate exponential autocorrelation')
279
280
281
282
283
284
285
286
287
288
289
290
291
292
            c = np.zeros(self.ndim) + np.nan

        self.convergence_diagnosticx.append(i - self.convergence_n/2.)
        self.convergence_diagnostic.append(list(c))

        if test:
            return i > n_cut * np.max(c)

    def test_GR_convergence(self, i, sampler, test=True, R=1.1):
        if self.convergence_windowed:
            s = sampler.chain[0, :, i-self.convergence_n+1:i+1, :]
        else:
            s = sampler.chain[0, :, :i+1, :]
        N = float(self.convergence_n)
293
294
        M = float(self.nwalkers)
        W = np.mean(np.var(s, axis=1), axis=0)
295
296
        per_walker_mean = np.mean(s, axis=1)
        mean = np.mean(per_walker_mean, axis=0)
297
298
        B = N / (M-1.) * np.sum((per_walker_mean-mean)**2, axis=0)
        Vhat = (N-1)/N * W + (M+1)/(M*N) * B
299
        c = np.sqrt(Vhat/W)
300
        self.convergence_diagnostic.append(c)
301
        self.convergence_diagnosticx.append(i - self.convergence_n/2.)
302

303
304
305
        if test and np.max(c) < R:
            return True
        else:
306
            return False
307
308
309
310

    def _test_convergence(self, i, sampler, **kwargs):
        if np.mod(i+1, self.convergence_n) == 0:
            return self._get_convergence_test(i, sampler, **kwargs)
311
        else:
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
            return False

    def _run_sampler_with_conv_test(self, sampler, p0, nprod=0, nburn=0):
        logging.info('Running {} burn-in steps with convergence testing'
                     .format(nburn))
        iterator = tqdm(sampler.sample(p0, iterations=nburn), total=nburn)
        for i, output in enumerate(iterator):
            if self._test_convergence(i, sampler, test=True,
                                      **self.convergence_kwargs):
                logging.info(
                    'Converged at {} before max number {} of steps reached'
                    .format(i, nburn))
                self.convergence_idx = i
                break
        iterator.close()
        logging.info('Running {} production steps'.format(nprod))
        j = nburn
        iterator = tqdm(sampler.sample(output[0], iterations=nprod),
                        total=nprod)
        for result in iterator:
            self._test_convergence(j, sampler, test=False,
                                   **self.convergence_kwargs)
            j += 1
        return sampler
336

337
    def _run_sampler(self, sampler, p0, nprod=0, nburn=0):
338
339
        if hasattr(self, 'convergence_n'):
            self._run_sampler_with_conv_test(sampler, p0, nprod, nburn)
340
341
342
343
        else:
            for result in tqdm(sampler.sample(p0, iterations=nburn+nprod),
                               total=nburn+nprod):
                pass
344

345
346
        self.mean_acceptance_fraction = np.mean(
            sampler.acceptance_fraction, axis=1)
347
        logging.info("Mean acceptance fraction: {}"
348
                     .format(self.mean_acceptance_fraction))
349
        if self.ntemps > 1:
350
            self.tswap_acceptance_fraction = sampler.tswap_acceptance_fraction
351
352
353
            logging.info("Tswap acceptance fraction: {}"
                         .format(sampler.tswap_acceptance_fraction))
        try:
354
            self.autocorr_time = sampler.get_autocorr_time(c=4)
355
            logging.info("Autocorrelation length: {}".format(
356
                self.autocorr_time))
357
        except emcee.autocorr.AutocorrError as e:
358
            self.autocorr_time = np.nan
359
360
361
362
363
            logging.warning(
                'Autocorrelation calculation failed with message {}'.format(e))

        return sampler

364
    def _estimate_run_time(self):
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
        """ Print the estimated run time

        Uses timing coefficients based on a Lenovo T460p Intel(R)
        Core(TM) i5-6300HQ CPU @ 2.30GHz.

        """
        # Todo: add option to time on a machine, and move coefficients to
        # ~/.pyfstat.conf
        if (type(self.theta_prior['Alpha']) == dict or
                type(self.theta_prior['Delta']) == dict):
            tau0S = 7.3e-5
            tau0LD = 4.2e-7
        else:
            tau0S = 5.0e-5
            tau0LD = 6.2e-8
380
        Nsfts = (self.maxStartTime - self.minStartTime) / 1800.
381
382
383
        numb_evals = np.sum(self.nsteps)*self.nwalkers*self.ntemps
        a = tau0S * numb_evals
        b = tau0LD * Nsfts * numb_evals
384
385
386
        logging.info('Estimated run-time = {} s = {:1.0f}:{:1.0f} m'.format(
            a+b, *divmod(a+b, 60)))

387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
    def run(self, proposal_scale_factor=2, create_plots=True, c=5, **kwargs):
        """ Run the MCMC simulatation

        Parameters
        ----------
        proposal_scale_factor: float
            The proposal scale factor used by the sampler, see Goodman & Weare
            (2010). If the acceptance fraction is too low, you can raise it by
            decreasing the a parameter; and if it is too high, you can reduce
            it by increasing the a parameter [Foreman-Mackay (2013)].
        create_plots: bool
            If true, save trace plots of the walkers
        c: int
            The minimum number of autocorrelation times needed to trust the
            result when estimating the autocorrelation time (see
            emcee.autocorr.integrated_time for further details. Default is 5
        **kwargs:
            Passed to _plot_walkers to control the figures

        """
407

408
        self.old_data_is_okay_to_use = self._check_old_data_is_okay_to_use()
409
410
411
        if self.old_data_is_okay_to_use is True:
            logging.warning('Using saved data from {}'.format(
                self.pickle_path))
412
            d = self.get_saved_data_dictionary()
413
414
415
            self.samples = d['samples']
            self.lnprobs = d['lnprobs']
            self.lnlikes = d['lnlikes']
416
            self.all_lnlikelihood = d['all_lnlikelihood']
417
418
            return

419
        self._initiate_search_object()
420
        self._estimate_run_time()
421
422
423
424

        sampler = emcee.PTSampler(
            self.ntemps, self.nwalkers, self.ndim, self.logl, self.logp,
            logpargs=(self.theta_prior, self.theta_keys, self.search),
425
            loglargs=(self.search,), betas=self.betas, a=proposal_scale_factor)
426

427
428
429
        p0 = self._generate_initial_p0()
        p0 = self._apply_corrections_to_p0(p0)
        self._check_initial_points(p0)
430
431
432
433

        ninit_steps = len(self.nsteps) - 2
        for j, n in enumerate(self.nsteps[:-2]):
            logging.info('Running {}/{} initialisation with {} steps'.format(
Gregory Ashton's avatar
Gregory Ashton committed
434
                j, ninit_steps, n))
435
            sampler = self._run_sampler(sampler, p0, nburn=n)
436
            if create_plots:
437
                fig, axes = self._plot_walkers(sampler,
438
439
                                               symbols=self.theta_symbols,
                                               **kwargs)
440
441
                fig.tight_layout()
                fig.savefig('{}/{}_init_{}_walkers.png'.format(
Gregory Ashton's avatar
Gregory Ashton committed
442
                    self.outdir, self.label, j))
443

444
445
446
            p0 = self._get_new_p0(sampler)
            p0 = self._apply_corrections_to_p0(p0)
            self._check_initial_points(p0)
447
448
            sampler.reset()

Gregory Ashton's avatar
Gregory Ashton committed
449
450
451
452
        if len(self.nsteps) > 1:
            nburn = self.nsteps[-2]
        else:
            nburn = 0
453
454
455
        nprod = self.nsteps[-1]
        logging.info('Running final burn and prod with {} steps'.format(
            nburn+nprod))
456
        sampler = self._run_sampler(sampler, p0, nburn=nburn, nprod=nprod)
457
        if create_plots:
458
            fig, axes = self._plot_walkers(sampler, symbols=self.theta_symbols,
459
                                           nprod=nprod, **kwargs)
460
461
            fig.tight_layout()
            fig.savefig('{}/{}_walkers.png'.format(self.outdir, self.label),
Gregory Ashton's avatar
Gregory Ashton committed
462
                        )
463
464
465
466

        samples = sampler.chain[0, :, nburn:, :].reshape((-1, self.ndim))
        lnprobs = sampler.lnprobability[0, :, nburn:].reshape((-1))
        lnlikes = sampler.lnlikelihood[0, :, nburn:].reshape((-1))
467
        all_lnlikelihood = sampler.lnlikelihood[:, :, nburn:]
468
469
470
        self.samples = samples
        self.lnprobs = lnprobs
        self.lnlikes = lnlikes
471
472
        self.all_lnlikelihood = all_lnlikelihood
        self._save_data(sampler, samples, lnprobs, lnlikes, all_lnlikelihood)
Gregory Ashton's avatar
Gregory Ashton committed
473
        return sampler
474

475
    def _get_rescale_multiplier_for_key(self, key):
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
        """ Get the rescale multiplier from the rescale_dictionary

        Can either be a float, a string (in which case it is interpretted as
        a attribute of the MCMCSearch class, e.g. minStartTime, or non-existent
        in which case 0 is returned
        """
        if key not in self.rescale_dictionary:
            return 1

        if 'multiplier' in self.rescale_dictionary[key]:
            val = self.rescale_dictionary[key]['multiplier']
            if type(val) == str:
                if hasattr(self, val):
                    multiplier = getattr(
                        self, self.rescale_dictionary[key]['multiplier'])
                else:
                    raise ValueError(
                        "multiplier {} not a class attribute".format(val))
            else:
                multiplier = val
        else:
            multiplier = 1
        return multiplier

500
    def _get_rescale_subtractor_for_key(self, key):
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
        """ Get the rescale subtractor from the rescale_dictionary

        Can either be a float, a string (in which case it is interpretted as
        a attribute of the MCMCSearch class, e.g. minStartTime, or non-existent
        in which case 0 is returned
        """
        if key not in self.rescale_dictionary:
            return 0

        if 'subtractor' in self.rescale_dictionary[key]:
            val = self.rescale_dictionary[key]['subtractor']
            if type(val) == str:
                if hasattr(self, val):
                    subtractor = getattr(
                        self, self.rescale_dictionary[key]['subtractor'])
                else:
                    raise ValueError(
                        "subtractor {} not a class attribute".format(val))
            else:
                subtractor = val
        else:
            subtractor = 0
        return subtractor

525
    def _scale_samples(self, samples, theta_keys):
526
        """ Scale the samples using the rescale_dictionary """
527
528
529
530
        for key in theta_keys:
            if key in self.rescale_dictionary:
                idx = theta_keys.index(key)
                s = samples[:, idx]
531
                subtractor = self._get_rescale_subtractor_for_key(key)
532
                s = s - subtractor
533
                multiplier = self._get_rescale_multiplier_for_key(key)
534
                s *= multiplier
535
536
                samples[:, idx] = s

537
538
        return samples

539
    def _get_labels(self):
540
        """ Combine the units, symbols and rescaling to give labels """
541

542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
        labels = []
        for key in self.theta_keys:
            label = None
            s = self.symbol_dictionary[key]
            s.replace('_{glitch}', r'_\textrm{glitch}')
            u = self.unit_dictionary[key]
            if key in self.rescale_dictionary:
                if 'symbol' in self.rescale_dictionary[key]:
                    s = self.rescale_dictionary[key]['symbol']
                if 'label' in self.rescale_dictionary[key]:
                    label = self.rescale_dictionary[key]['label']
                if 'unit' in self.rescale_dictionary[key]:
                    u = self.rescale_dictionary[key]['unit']
            if label is None:
                label = '{} \n [{}]'.format(s, u)
            labels.append(label)
        return labels
559

560
561
    def plot_corner(self, figsize=(7, 7), add_prior=False, nstds=None,
                    label_offset=0.4, dpi=300, rc_context={},
562
                    tglitch_ratio=False, fig_and_axes=None, save_fig=True,
563
                    **kwargs):
564
565
566
567
568
569
570
571
572
        """ Generate a corner plot of the posterior

        Using the `corner` package (https://pypi.python.org/pypi/corner/),
        generate estimates of the posterior from the production samples.

        Parameters
        ----------
        figsize: tuple (7, 7)
            Figure size in inches (passed to plt.subplots)
573
574
575
        add_prior: bool, str
            If true, plot the prior as a red line. If 'full' then for uniform
            priors plot the full extent of the prior.
576
577
578
579
580
581
582
583
584
585
586
587
588
589
        nstds: float
            The number of standard deviations to plot centered on the mean
        label_offset: float
            Offset the labels from the plot: useful to precent overlapping the
            tick labels with the axis labels
        dpi: int
            Passed to plt.savefig
        rc_context: dict
            Dictionary of rc values to set while generating the figure (see
            matplotlib rc for more details)
        tglitch_ratio: bool
            If true, and tglitch is a parameter, plot posteriors as the
            fractional time at which the glitch occurs instead of the actual
            time
590
591
592
593
594
        fig_and_axes: tuple
            fig and axes to plot on, the axes must be of the right shape,
            namely (ndim, ndim)
        save_fig: bool
            If true, save the figure, else return the fig, axes
595

596
        Note: kwargs are passed on to corner.corner
597
598

        """
599

600
601
602
603
        if 'truths' in kwargs and len(kwargs['truths']) != self.ndim:
            logging.warning('len(Truths) != ndim, Truths will be ignored')
            kwargs['truths'] = None

Gregory Ashton's avatar
Gregory Ashton committed
604
605
        if self.ndim < 2:
            with plt.rc_context(rc_context):
606
607
608
609
                if fig_and_axes is None:
                    fig, ax = plt.subplots(figsize=figsize)
                else:
                    fig, ax = fig_and_axes
Gregory Ashton's avatar
Gregory Ashton committed
610
611
612
613
614
615
616
                ax.hist(self.samples, bins=50, histtype='stepfilled')
                ax.set_xlabel(self.theta_symbols[0])

            fig.savefig('{}/{}_corner.png'.format(
                self.outdir, self.label), dpi=dpi)
            return

617
        with plt.rc_context(rc_context):
618
619
620
621
622
            if fig_and_axes is None:
                fig, axes = plt.subplots(self.ndim, self.ndim,
                                         figsize=figsize)
            else:
                fig, axes = fig_and_axes
623
624

            samples_plt = copy.copy(self.samples)
625
            labels = self._get_labels()
626

627
            samples_plt = self._scale_samples(samples_plt, self.theta_keys)
628
629
630
631
632

            if tglitch_ratio:
                for j, k in enumerate(self.theta_keys):
                    if k == 'tglitch':
                        s = samples_plt[:, j]
633
634
635
                        samples_plt[:, j] = (
                            s - self.minStartTime)/(
                                self.maxStartTime - self.minStartTime)
636
                        labels[j] = r'$R_{\textrm{glitch}}$'
637
638
639
640
641
642
643

            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))
644
645
            elif 'range' in kwargs:
                _range = kwargs.pop('range')
646
647
648
            else:
                _range = None

649
650
651
652
            hist_kwargs = kwargs.pop('hist_kwargs', dict())
            if 'normed' not in hist_kwargs:
                hist_kwargs['normed'] = True

653
            fig_triangle = corner.corner(samples_plt,
654
                                         labels=labels,
655
656
657
658
659
660
661
662
663
                                         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,
664
                                         hist_kwargs=hist_kwargs,
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
                                         **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:
681
                self._add_prior_to_corner(axes, self.samples, add_prior)
682

683
684
685
686
687
            if save_fig:
                fig_triangle.savefig('{}/{}_corner.png'.format(
                    self.outdir, self.label), dpi=dpi)
            else:
                return fig, axes
688

689
    def _add_prior_to_corner(self, axes, samples, add_prior):
690
691
692
        for i, key in enumerate(self.theta_keys):
            ax = axes[i][i]
            s = samples[:, i]
693
694
695
696
697
698
699
700
701
702
            lnprior = self._generic_lnprior(**self.theta_prior[key])
            if add_prior == 'full' and self.theta_prior[key]['type'] == 'unif':
                lower = self.theta_prior[key]['lower']
                upper = self.theta_prior[key]['upper']
                r = upper-lower
                xlim = [lower-0.05*r, upper+0.05*r]
                x = np.linspace(xlim[0], xlim[1], 1000)
            else:
                xlim = ax.get_xlim()
                x = np.linspace(s.min(), s.max(), 1000)
703
704
            multiplier = self._get_rescale_multiplier_for_key(key)
            subtractor = self._get_rescale_subtractor_for_key(key)
705
706
707
708
709
710
711
712
            ax.plot((x-subtractor)*multiplier,
                    [np.exp(lnprior(xi)) for xi in x], '-C3',
                    label='prior')

            for j in range(i, self.ndim):
                axes[j][i].set_xlim(xlim[0], xlim[1])
            for k in range(0, i):
                axes[i][k].set_ylim(xlim[0], xlim[1])
713

714
715
716
717
718
719
720
721
    def plot_prior_posterior(self, normal_stds=2):
        """ Plot the posterior in the context of the prior """
        fig, axes = plt.subplots(nrows=self.ndim, figsize=(8, 4*self.ndim))
        N = 1000
        from scipy.stats import gaussian_kde

        for i, (ax, key) in enumerate(zip(axes, self.theta_keys)):
            prior_dict = self.theta_prior[key]
722
            prior_func = self._generic_lnprior(**prior_dict)
723
724
725
726
727
            if prior_dict['type'] == 'unif':
                x = np.linspace(prior_dict['lower'], prior_dict['upper'], N)
                prior = prior_func(x)
                prior[0] = 0
                prior[-1] = 0
Gregory Ashton's avatar
Gregory Ashton committed
728
729
730
731
732
            elif prior_dict['type'] == 'log10unif':
                upper = prior_dict['log10upper']
                lower = prior_dict['log10lower']
                x = np.linspace(lower, upper, N)
                prior = [prior_func(xi) for xi in x]
733
734
735
736
737
            elif prior_dict['type'] == 'norm':
                lower = prior_dict['loc'] - normal_stds * prior_dict['scale']
                upper = prior_dict['loc'] + normal_stds * prior_dict['scale']
                x = np.linspace(lower, upper, N)
                prior = prior_func(x)
738
739
740
741
742
            elif prior_dict['type'] == 'halfnorm':
                lower = prior_dict['loc']
                upper = prior_dict['loc'] + normal_stds * prior_dict['scale']
                x = np.linspace(lower, upper, N)
                prior = [prior_func(xi) for xi in x]
Gregory Ashton's avatar
Gregory Ashton committed
743
744
745
746
747
            elif prior_dict['type'] == 'neghalfnorm':
                upper = prior_dict['loc']
                lower = prior_dict['loc'] - normal_stds * prior_dict['scale']
                x = np.linspace(lower, upper, N)
                prior = [prior_func(xi) for xi in x]
748
749
750
            else:
                raise ValueError('Not implemented for prior type {}'.format(
                    prior_dict['type']))
751
            priorln = ax.plot(x, prior, 'C3', label='prior')
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
            ax.set_xlabel(self.theta_symbols[i])

            s = self.samples[:, i]
            while len(s) > 10**4:
                # random downsample to avoid slow calculation of kde
                s = np.random.choice(s, size=int(len(s)/2.))
            kde = gaussian_kde(s)
            ax2 = ax.twinx()
            postln = ax2.plot(x, kde.pdf(x), 'k', label='posterior')
            ax2.set_yticklabels([])
            ax.set_yticklabels([])

        lns = priorln + postln
        labs = [l.get_label() for l in lns]
        axes[0].legend(lns, labs, loc=1, framealpha=0.8)

        fig.savefig('{}/{}_prior_posterior.png'.format(
            self.outdir, self.label))

771
    def plot_cumulative_max(self, **kwargs):
772
773
774
775
        """ Plot the cumulative twoF for the maximum posterior estimate

        See the pyfstat.core.plot_twoF_cumulative function for further details
        """
Gregory Ashton's avatar
Gregory Ashton committed
776
777
778
779
        d, maxtwoF = self.get_max_twoF()
        for key, val in self.theta_prior.iteritems():
            if key not in d:
                d[key] = val
780
781

        if hasattr(self, 'search') is False:
782
            self._initiate_search_object()
783
784
785
        if self.binary is False:
            self.search.plot_twoF_cumulative(
                self.label, self.outdir, F0=d['F0'], F1=d['F1'], F2=d['F2'],
786
                Alpha=d['Alpha'], Delta=d['Delta'],
787
                tstart=self.minStartTime, tend=self.maxStartTime,
788
                **kwargs)
789
790
791
792
793
        else:
            self.search.plot_twoF_cumulative(
                self.label, self.outdir, F0=d['F0'], F1=d['F1'], F2=d['F2'],
                Alpha=d['Alpha'], Delta=d['Delta'], asini=d['asini'],
                period=d['period'], ecc=d['ecc'], argp=d['argp'], tp=d['argp'],
794
                tstart=self.minStartTime, tend=self.maxStartTime, **kwargs)
Gregory Ashton's avatar
Gregory Ashton committed
795

796
    def _generic_lnprior(self, **kwargs):
797
798
799
800
801
802
803
804
805
        """ Return a lambda function of the pdf

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

        """

Gregory Ashton's avatar
Gregory Ashton committed
806
        def log_of_unif(x, a, b):
807
808
809
810
811
812
813
814
815
816
817
818
819
            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

Gregory Ashton's avatar
Gregory Ashton committed
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
        def log_of_log10unif(x, log10lower, log10upper):
            log10x = np.log10(x)
            above = log10x < log10upper
            below = log10x > log10lower
            if type(above) is not np.ndarray:
                if above and below:
                    return -np.log(x*np.log(10)*(log10upper-log10lower))
                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(x*np.log(10)*(log10upper-log10lower))
                return p

        def log_of_halfnorm(x, loc, scale):
836
            if x < loc:
837
838
839
840
841
842
843
844
845
846
847
848
849
850
                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':
Gregory Ashton's avatar
Gregory Ashton committed
851
852
853
854
            return lambda x: log_of_unif(x, kwargs['lower'], kwargs['upper'])
        if kwargs['type'] == 'log10unif':
            return lambda x: log_of_log10unif(
                x, kwargs['log10lower'], kwargs['log10upper'])
855
        elif kwargs['type'] == 'halfnorm':
Gregory Ashton's avatar
Gregory Ashton committed
856
            return lambda x: log_of_halfnorm(x, kwargs['loc'], kwargs['scale'])
857
        elif kwargs['type'] == 'neghalfnorm':
Gregory Ashton's avatar
Gregory Ashton committed
858
859
            return lambda x: log_of_halfnorm(
                -x, kwargs['loc'], kwargs['scale'])
860
861
862
863
864
865
866
        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")

867
    def _generate_rv(self, **kwargs):
868
869
870
        dist_type = kwargs.pop('type')
        if dist_type == "unif":
            return np.random.uniform(low=kwargs['lower'], high=kwargs['upper'])
Gregory Ashton's avatar
Gregory Ashton committed
871
872
873
        if dist_type == "log10unif":
            return 10**(np.random.uniform(low=kwargs['log10lower'],
                                          high=kwargs['log10upper']))
874
875
876
877
878
        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']))
879
880
881
        if dist_type == "neghalfnorm":
            return -1 * np.abs(np.random.normal(loc=kwargs['loc'],
                                                scale=kwargs['scale']))
882
883
884
885
886
887
        if dist_type == "lognorm":
            return np.random.lognormal(
                mean=kwargs['loc'], sigma=kwargs['scale'])
        else:
            raise ValueError("dist_type {} unknown".format(dist_type))

888
    def _plot_walkers(self, sampler, symbols=None, alpha=0.8, color="k",
889
890
                      temp=0, lw=0.1, nprod=0, add_det_stat_burnin=False,
                      fig=None, axes=None, xoffset=0, plot_det_stat=False,
891
                      context='ggplot', subtractions=None, labelpad=0.05):
892
893
        """ Plot all the chains from a sampler """

894
895
896
897
898
        if context not in plt.style.available:
            raise ValueError((
                'The requested context {} is not available; please select a'
                ' context from `plt.style.available`').format(context))

899
900
901
        if np.ndim(axes) > 1:
            axes = axes.flatten()

902
903
904
905
906
907
908
909
910
911
912
913
914
        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, :, :, :]

915
916
        if subtractions is None:
            subtractions = [0 for i in range(ndim)]
917
918
919
        else:
            if len(subtractions) != self.ndim:
                raise ValueError('subtractions must be of length ndim')
920

921
922
923
924
        if plot_det_stat:
            extra_subplots = 1
        else:
            extra_subplots = 0
925
        with plt.style.context((context)):
Gregory Ashton's avatar
Gregory Ashton committed
926
            plt.rcParams['text.usetex'] = True
Gregory Ashton's avatar
Gregory Ashton committed
927
            if fig is None and axes is None:
928
                fig = plt.figure(figsize=(4, 3.0*ndim))
929
930
                ax = fig.add_subplot(ndim+extra_subplots, 1, 1)
                axes = [ax] + [fig.add_subplot(ndim+extra_subplots, 1, i)
Gregory Ashton's avatar
Gregory Ashton committed
931
                               for i in range(2, ndim+1)]
932

Gregory Ashton's avatar
Gregory Ashton committed
933
            idxs = np.arange(chain.shape[1])
934
935
936
937
938
            burnin_idx = chain.shape[1] - nprod
            if hasattr(self, 'convergence_idx'):
                convergence_idx = self.convergence_idx
            else:
                convergence_idx = burnin_idx
939
940
            if ndim > 1:
                for i in range(ndim):
941
                    axes[i].ticklabel_format(useOffset=False, axis='y')
Gregory Ashton's avatar
Gregory Ashton committed
942
                    cs = chain[:, :, i].T
943
                    if burnin_idx > 0:
944
945
                        axes[i].plot(xoffset+idxs[:convergence_idx+1],
                                     cs[:convergence_idx+1]-subtractions[i],
946
                                     color="C3", alpha=alpha,
Gregory Ashton's avatar
Gregory Ashton committed
947
                                     lw=lw)
948
                        axes[i].axvline(xoffset+convergence_idx,
949
                                        color='k', ls='--', lw=0.25)
950
951
                    axes[i].plot(xoffset+idxs[burnin_idx:],
                                 cs[burnin_idx:]-subtractions[i],
Gregory Ashton's avatar
Gregory Ashton committed
952
                                 color="k", alpha=alpha, lw=lw)
Gregory Ashton's avatar
Gregory Ashton committed
953
954

                    axes[i].set_xlim(0, xoffset+idxs[-1])
955
                    if symbols:
956
                        if subtractions[i] == 0:
957
                            axes[i].set_ylabel(symbols[i], labelpad=labelpad)
958
959
                        else:
                            axes[i].set_ylabel(
960
961
                                symbols[i]+'$-$'+symbols[i]+'$_0$',
                                labelpad=labelpad)
962

963
964
                    if hasattr(self, 'convergence_diagnostic'):
                        ax = axes[i].twinx()
965
966
                        axes[i].set_zorder(ax.get_zorder()+1)
                        axes[i].patch.set_visible(False)
967
968
                        c_x = np.array(self.convergence_diagnosticx)
                        c_y = np.array(self.convergence_diagnostic)
969
                        break_idx = np.argmin(np.abs(c_x - burnin_idx))
970
971
972
973
                        ax.plot(c_x[:break_idx], c_y[:break_idx, i], '-C0',
                                zorder=-10)
                        ax.plot(c_x[break_idx:], c_y[break_idx:, i], '-C0',
                                zorder=-10)
974
975
976
977
                        if self.convergence_test_type == 'autocorr':
                            ax.set_ylabel(r'$\tau_\mathrm{exp}$')
                        elif self.convergence_test_type == 'GR':
                            ax.set_ylabel('PSRF')
978
                        ax.ticklabel_format(useOffset=False)
979
            else:
Gregory Ashton's avatar
Gregory Ashton committed
980
                axes[0].ticklabel_format(useOffset=False, axis='y')
Gregory Ashton's avatar
Gregory Ashton committed
981
                cs = chain[:, :, temp].T
Gregory Ashton's avatar
Gregory Ashton committed
982
983
                if burnin_idx:
                    axes[0].plot(idxs[:burnin_idx], cs[:burnin_idx],
984
                                 color="C3", alpha=alpha, lw=lw)
Gregory Ashton's avatar
Gregory Ashton committed
985
986
987
                axes[0].plot(idxs[burnin_idx:], cs[burnin_idx:], color="k",
                             alpha=alpha, lw=lw)
                if symbols:
988
                    axes[0].set_ylabel(symbols[0], labelpad=labelpad)
989

Gregory Ashton's avatar
Gregory Ashton committed
990
991
            axes[-1].set_xlabel(r'$\textrm{Number of steps}$', labelpad=0.2)

992
            if plot_det_stat:
993
994
995
                if len(axes) == ndim:
                    axes.append(fig.add_subplot(ndim+1, 1, ndim+1))

996
997
998
                lnl = sampler.lnlikelihood[temp, :, :]
                if burnin_idx and add_det_stat_burnin:
                    burn_in_vals = lnl[:, :burnin_idx].flatten()
999
                    try:
1000
                        twoF_burnin = (burn_in_vals[~np.isnan(burn_in_vals)]