mcmc_based_searches.py 92 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
28
29
30
31
32
33
34

    Parameters
    ----------
    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.
    tref, minStartTime, maxStartTime: int
35
36
37
38
39
40
41
        GPS seconds of the reference time, start time and end time. While tref
        is requirede, minStartTime and maxStartTime default to None in which
        case all available data is used.
    label, outdir: str
        A label and output directory (optional, defaults is `'data'`) to
        name files
    sftfilepattern: str, optional
Gregory Ashton's avatar
Gregory Ashton committed
42
43
        Pattern to match SFTs using wildcards (*?) and ranges [0-9];
        mutiple patterns can be given separated by colons.
44
    detectors: str, optional
Gregory Ashton's avatar
Gregory Ashton committed
45
46
        Two character reference to the detectors to use, specify None for no
        contraint and comma separate for multiple references.
47
    nsteps: list (2,), optional
48
49
50
        Number of burn-in and production steps to take, [nburn, nprod]. See
        `pyfstat.MCMCSearch.setup_initialisation()` for details on adding
        initialisation steps.
51
    nwalkers, ntemps: int, optional
52
53
        The number of walkers and temperates to use in the parallel
        tempered PTSampler.
54
    log10beta_min float < 0, optional
55
56
57
        The  log_10(beta) value, if given the set of betas passed to PTSampler
        are generated from `np.logspace(0, log10beta_min, ntemps)` (given
        in descending order to emcee).
58
    theta_initial: dict, array, optional
59
60
        A dictionary of distribution about which to distribute the
        initial walkers about
61
    rhohatmax: float, optional
62
63
64
        Upper bound for the SNR scale parameter (required to normalise the
        Bayes factor) - this needs to be carefully set when using the
        evidence.
65
    binary: bool, optional
66
        If true, search over binary parameters
67
    BSGL: bool, optional
Gregory Ashton's avatar
Gregory Ashton committed
68
        If true, use the BSGL statistic
69
    SSBPrec: int, optional
Gregory Ashton's avatar
Gregory Ashton committed
70
        SSBPrec (SSB precision) to use when calling ComputeFstat
71
    minCoverFreq, maxCoverFreq: float, optional
72
73
        Minimum and maximum instantaneous frequency which will be covered
        over the SFT time span as passed to CreateFstatInput
74
    injectSources: dict, optional
Gregory Ashton's avatar
Gregory Ashton committed
75
76
        If given, inject these properties into the SFT files before running
        the search
77
    assumeSqrtSX: float, optional
Gregory Ashton's avatar
Gregory Ashton committed
78
        Don't estimate noise-floors, but assume (stationary) per-IFO sqrt{SX}
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93

    Attributes
    ----------
    symbol_dictionary: dict
        Key, val pairs of the parameters (i.e. `F0`, `F1`), to Latex math
        symbols for plots
    unit_dictionary: dict
        Key, val pairs of the parameters (i.e. `F0`, `F1`), and the
        units (i.e. `Hz`)
    transform_dictionary: dict
        Key, val pairs of the parameters (i.e. `F0`, `F1`), where the key is
        itself a dictionary which can item `multiplier`, `subtractor`, or
        `unit` by which to transform by and update the units.

    """
94
95

    symbol_dictionary = dict(
96
        F0='$f$', F1='$\dot{f}$', F2='$\ddot{f}$', Alpha=r'$\alpha$',
97
98
        Delta='$\delta$', asini='asini', period='P', ecc='ecc', tp='tp',
        argp='argp')
99
    unit_dictionary = dict(
100
101
        F0='Hz', F1='Hz/s', F2='Hz/s$^2$', Alpha=r'rad', Delta='rad',
        asini='', period='s', ecc='', tp='', argp='')
102
    transform_dictionary = {}
103

Gregory Ashton's avatar
Gregory Ashton committed
104
    @helper_functions.initializer
105
106
107
    def __init__(self, theta_prior, tref, label, outdir='data',
                 minStartTime=None, maxStartTime=None, sftfilepattern=None,
                 detectors=None, nsteps=[100, 100], nwalkers=100, ntemps=1,
108
                 log10beta_min=-5, theta_initial=None,
109
                 rhohatmax=1000, binary=False, BSGL=False,
Gregory Ashton's avatar
Gregory Ashton committed
110
                 SSBprec=None, minCoverFreq=None, maxCoverFreq=None,
111
                 injectSources=None, assumeSqrtSX=None):
112

Gregory Ashton's avatar
Gregory Ashton committed
113
114
        if os.path.isdir(outdir) is False:
            os.mkdir(outdir)
115
        self._add_log_file()
116
        logging.info('Set-up MCMC search for model {}'.format(self.label))
117
118
        if sftfilepattern:
            logging.info('Using data {}'.format(self.sftfilepattern))
119
        else:
120
            logging.info('No sftfilepattern given')
121
122
        if injectSources:
            logging.info('Inject sources: {}'.format(injectSources))
123
        self.pickle_path = '{}/{}_saved_data.p'.format(self.outdir, self.label)
124
        self._unpack_input_theta()
125
        self.ndim = len(self.theta_keys)
126
127
        if self.log10beta_min:
            self.betas = np.logspace(0, self.log10beta_min, self.ntemps)
128
129
        else:
            self.betas = None
130

131
132
133
        if args.clean and os.path.isfile(self.pickle_path):
            os.rename(self.pickle_path, self.pickle_path+".old")

134
        self._set_likelihoodcoef()
135
        self._log_input()
136
137
138

    def _set_likelihoodcoef(self):
        self.likelihoodcoef = np.log(70./self.rhohatmax**4)
139

140
    def _log_input(self):
141
        logging.info('theta_prior = {}'.format(self.theta_prior))
142
        logging.info('nwalkers={}'.format(self.nwalkers))
143
144
        logging.info('nsteps = {}'.format(self.nsteps))
        logging.info('ntemps = {}'.format(self.ntemps))
145
146
        logging.info('log10beta_min = {}'.format(
            self.log10beta_min))
147

148
    def _initiate_search_object(self):
149
        logging.info('Setting up search object')
150
        self.search = core.ComputeFstat(
151
            tref=self.tref, sftfilepattern=self.sftfilepattern,
152
            minCoverFreq=self.minCoverFreq, maxCoverFreq=self.maxCoverFreq,
153
            detectors=self.detectors, BSGL=self.BSGL, transient=False,
154
            minStartTime=self.minStartTime, maxStartTime=self.maxStartTime,
155
            binary=self.binary, injectSources=self.injectSources,
156
            assumeSqrtSX=self.assumeSqrtSX, SSBprec=self.SSBprec)
157
158
159
160
        if self.minStartTime is None:
            self.minStartTime = self.search.minStartTime
        if self.maxStartTime is None:
            self.maxStartTime = self.search.maxStartTime
161
162

    def logp(self, theta_vals, theta_prior, theta_keys, search):
163
        H = [self._generic_lnprior(**theta_prior[key])(p) for p, key in
164
165
166
167
168
169
             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]
170
171
172
        twoF = search.get_fullycoherent_twoF(
            self.minStartTime, self.maxStartTime, *self.fixed_theta)
        return twoF/2.0 + self.likelihoodcoef
173

174
    def _unpack_input_theta(self):
175
        full_theta_keys = ['F0', 'F1', 'F2', 'Alpha', 'Delta']
176
177
178
        if self.binary:
            full_theta_keys += [
                'asini', 'period', 'ecc', 'tp', 'argp']
179
180
        full_theta_keys_copy = copy.copy(full_theta_keys)

181
182
        full_theta_symbols = ['$f$', '$\dot{f}$', '$\ddot{f}$', r'$\alpha$',
                              r'$\delta$']
183
184
        if self.binary:
            full_theta_symbols += [
185
                'asini', 'period', 'ecc', 'tp', 'argp']
186

187
188
        self.theta_keys = []
        fixed_theta_dict = {}
189
        for key, val in self.theta_prior.iteritems():
190
191
            if type(val) is dict:
                fixed_theta_dict[key] = 0
Gregory Ashton's avatar
Gregory Ashton committed
192
                self.theta_keys.append(key)
193
194
195
196
197
198
            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
199
            full_theta_keys_copy.pop(full_theta_keys_copy.index(key))
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214

        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]

215
    def _check_initial_points(self, p0):
216
217
218
219
220
221
222
223
224
225
226
227
228
        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))

229
                p0 = self._generate_new_p0_to_fix_initial_points(
230
231
                    p0, nt, initial_priors)

232
    def _generate_new_p0_to_fix_initial_points(self, p0, nt, initial_priors):
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
        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
252

253
254
    def setup_burnin_convergence_testing(
            self, n=10, test_type='autocorr', windowed=False, **kwargs):
255
        """ Set up convergence testing during the MCMC simulation
256
257
258

        Parameters
        ----------
259
260
261
262
263
264
265
266
267
        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`
268
269
270
271
        **kwargs:
            Passed to either `_test_autocorr_convergence()` or
            `_test_GR_convergence()` depending on `test_type`.

272
        """
273
        logging.info('Setting up convergence testing')
274
275
276
277
        self.convergence_n = n
        self.convergence_windowed = windowed
        self.convergence_test_type = test_type
        self.convergence_kwargs = kwargs
278
279
        self.convergence_diagnostic = []
        self.convergence_diagnosticx = []
280
        if test_type in ['autocorr']:
281
            self._get_convergence_test = self._test_autocorr_convergence
282
        elif test_type in ['GR']:
283
            self._get_convergence_test = self._test_GR_convergence
284
285
286
        else:
            raise ValueError('test_type {} not understood'.format(test_type))

287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
    def setup_initialisation(self, nburn0, scatter_val=1e-10):
        """ Add an initialisation step to the MCMC run

        If called prior to `run()`, adds an intial step in which the MCMC
        simulation is run for `nburn0` steps. After this, the MCMC simulation
        continues in the usual manner (i.e. for nburn and nprod steps), but the
        walkers are reset scattered around the maximum likelihood position
        of the initialisation step.

        Parameters
        ----------
        nburn0: int
            Number of initialisation steps to take
        scatter_val: float
            Relative number to scatter walkers around the maximum likelihood
            position after the initialisation step

        """

        logging.info('Setting up initialisation with nburn0={}, scatter_val={}'
                     .format(nburn0, scatter_val))
        self.nsteps = [nburn0] + self.nsteps
        self.scatter_val = scatter_val

311
    def _test_autocorr_convergence(self, i, sampler, test=True, n_cut=5):
312
313
314
315
316
317
318
319
320
321
322
        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
323
324
325
326
            logging.info('Failed to calculate exponential autocorrelation')
            c = np.zeros(self.ndim) + np.nan
        except AttributeError:
            logging.info('Unable to calculate exponential autocorrelation')
327
328
329
330
331
332
333
334
            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)

335
    def _test_GR_convergence(self, i, sampler, test=True, R=1.1):
336
337
338
339
340
        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)
341
342
        M = float(self.nwalkers)
        W = np.mean(np.var(s, axis=1), axis=0)
343
344
        per_walker_mean = np.mean(s, axis=1)
        mean = np.mean(per_walker_mean, axis=0)
345
346
        B = N / (M-1.) * np.sum((per_walker_mean-mean)**2, axis=0)
        Vhat = (N-1)/N * W + (M+1)/(M*N) * B
347
        c = np.sqrt(Vhat/W)
348
        self.convergence_diagnostic.append(c)
349
        self.convergence_diagnosticx.append(i - self.convergence_n/2.)
350

351
352
353
        if test and np.max(c) < R:
            return True
        else:
354
            return False
355
356
357
358

    def _test_convergence(self, i, sampler, **kwargs):
        if np.mod(i+1, self.convergence_n) == 0:
            return self._get_convergence_test(i, sampler, **kwargs)
359
        else:
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
            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
384

385
    def _run_sampler(self, sampler, p0, nprod=0, nburn=0):
386
387
        if hasattr(self, 'convergence_n'):
            self._run_sampler_with_conv_test(sampler, p0, nprod, nburn)
388
389
390
391
        else:
            for result in tqdm(sampler.sample(p0, iterations=nburn+nprod),
                               total=nburn+nprod):
                pass
392

393
394
        self.mean_acceptance_fraction = np.mean(
            sampler.acceptance_fraction, axis=1)
395
        logging.info("Mean acceptance fraction: {}"
396
                     .format(self.mean_acceptance_fraction))
397
        if self.ntemps > 1:
398
            self.tswap_acceptance_fraction = sampler.tswap_acceptance_fraction
399
400
401
            logging.info("Tswap acceptance fraction: {}"
                         .format(sampler.tswap_acceptance_fraction))
        try:
402
            self.autocorr_time = sampler.get_autocorr_time(c=4)
403
            logging.info("Autocorrelation length: {}".format(
404
                self.autocorr_time))
405
        except emcee.autocorr.AutocorrError as e:
406
            self.autocorr_time = np.nan
407
408
409
410
411
            logging.warning(
                'Autocorrelation calculation failed with message {}'.format(e))

        return sampler

412
    def _estimate_run_time(self):
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
        """ 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
428
        Nsfts = (self.maxStartTime - self.minStartTime) / 1800.
429
430
431
        numb_evals = np.sum(self.nsteps)*self.nwalkers*self.ntemps
        a = tau0S * numb_evals
        b = tau0LD * Nsfts * numb_evals
432
433
434
        logging.info('Estimated run-time = {} s = {:1.0f}:{:1.0f} m'.format(
            a+b, *divmod(a+b, 60)))

435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
    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

454
455
456
457
458
        Returns
        -------
        sampler: emcee.ptsampler.PTSampler
            The emcee ptsampler object

459
        """
460

461
        self.old_data_is_okay_to_use = self._check_old_data_is_okay_to_use()
462
463
464
        if self.old_data_is_okay_to_use is True:
            logging.warning('Using saved data from {}'.format(
                self.pickle_path))
465
            d = self.get_saved_data_dictionary()
466
467
468
            self.samples = d['samples']
            self.lnprobs = d['lnprobs']
            self.lnlikes = d['lnlikes']
469
            self.all_lnlikelihood = d['all_lnlikelihood']
470
471
            return

472
        self._initiate_search_object()
473
        self._estimate_run_time()
474
475
476
477

        sampler = emcee.PTSampler(
            self.ntemps, self.nwalkers, self.ndim, self.logl, self.logp,
            logpargs=(self.theta_prior, self.theta_keys, self.search),
478
            loglargs=(self.search,), betas=self.betas, a=proposal_scale_factor)
479

480
481
482
        p0 = self._generate_initial_p0()
        p0 = self._apply_corrections_to_p0(p0)
        self._check_initial_points(p0)
483

484
        # Run initialisation steps if required
485
486
487
        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
488
                j, ninit_steps, n))
489
            sampler = self._run_sampler(sampler, p0, nburn=n)
490
            if create_plots:
491
                fig, axes = self._plot_walkers(sampler,
492
493
                                               symbols=self.theta_symbols,
                                               **kwargs)
494
495
                fig.tight_layout()
                fig.savefig('{}/{}_init_{}_walkers.png'.format(
Gregory Ashton's avatar
Gregory Ashton committed
496
                    self.outdir, self.label, j))
497

498
499
500
            p0 = self._get_new_p0(sampler)
            p0 = self._apply_corrections_to_p0(p0)
            self._check_initial_points(p0)
501
502
            sampler.reset()

Gregory Ashton's avatar
Gregory Ashton committed
503
504
505
506
        if len(self.nsteps) > 1:
            nburn = self.nsteps[-2]
        else:
            nburn = 0
507
508
509
        nprod = self.nsteps[-1]
        logging.info('Running final burn and prod with {} steps'.format(
            nburn+nprod))
510
        sampler = self._run_sampler(sampler, p0, nburn=nburn, nprod=nprod)
511
        if create_plots:
512
            fig, axes = self._plot_walkers(sampler, symbols=self.theta_symbols,
513
                                           nprod=nprod, **kwargs)
514
515
            fig.tight_layout()
            fig.savefig('{}/{}_walkers.png'.format(self.outdir, self.label),
Gregory Ashton's avatar
Gregory Ashton committed
516
                        )
517
518
519
520

        samples = sampler.chain[0, :, nburn:, :].reshape((-1, self.ndim))
        lnprobs = sampler.lnprobability[0, :, nburn:].reshape((-1))
        lnlikes = sampler.lnlikelihood[0, :, nburn:].reshape((-1))
521
        all_lnlikelihood = sampler.lnlikelihood[:, :, nburn:]
522
523
524
        self.samples = samples
        self.lnprobs = lnprobs
        self.lnlikes = lnlikes
525
526
        self.all_lnlikelihood = all_lnlikelihood
        self._save_data(sampler, samples, lnprobs, lnlikes, all_lnlikelihood)
Gregory Ashton's avatar
Gregory Ashton committed
527
        return sampler
528

529
    def _get_rescale_multiplier_for_key(self, key):
530
        """ Get the rescale multiplier from the transform_dictionary
531
532
533
534
535

        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
        """
536
        if key not in self.transform_dictionary:
537
538
            return 1

539
540
        if 'multiplier' in self.transform_dictionary[key]:
            val = self.transform_dictionary[key]['multiplier']
541
542
543
            if type(val) == str:
                if hasattr(self, val):
                    multiplier = getattr(
544
                        self, self.transform_dictionary[key]['multiplier'])
545
546
547
548
549
550
551
552
553
                else:
                    raise ValueError(
                        "multiplier {} not a class attribute".format(val))
            else:
                multiplier = val
        else:
            multiplier = 1
        return multiplier

554
    def _get_rescale_subtractor_for_key(self, key):
555
        """ Get the rescale subtractor from the transform_dictionary
556
557
558
559
560

        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
        """
561
        if key not in self.transform_dictionary:
562
563
            return 0

564
565
        if 'subtractor' in self.transform_dictionary[key]:
            val = self.transform_dictionary[key]['subtractor']
566
567
568
            if type(val) == str:
                if hasattr(self, val):
                    subtractor = getattr(
569
                        self, self.transform_dictionary[key]['subtractor'])
570
571
572
573
574
575
576
577
578
                else:
                    raise ValueError(
                        "subtractor {} not a class attribute".format(val))
            else:
                subtractor = val
        else:
            subtractor = 0
        return subtractor

579
    def _scale_samples(self, samples, theta_keys):
580
        """ Scale the samples using the transform_dictionary """
581
        for key in theta_keys:
582
            if key in self.transform_dictionary:
583
584
                idx = theta_keys.index(key)
                s = samples[:, idx]
585
                subtractor = self._get_rescale_subtractor_for_key(key)
586
                s = s - subtractor
587
                multiplier = self._get_rescale_multiplier_for_key(key)
588
                s *= multiplier
589
590
                samples[:, idx] = s

591
592
        return samples

593
    def _get_labels(self):
594
        """ Combine the units, symbols and rescaling to give labels """
595

596
597
598
599
600
601
        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]
602
603
604
605
606
607
608
            if key in self.transform_dictionary:
                if 'symbol' in self.transform_dictionary[key]:
                    s = self.transform_dictionary[key]['symbol']
                if 'label' in self.transform_dictionary[key]:
                    label = self.transform_dictionary[key]['label']
                if 'unit' in self.transform_dictionary[key]:
                    u = self.transform_dictionary[key]['unit']
609
610
611
612
            if label is None:
                label = '{} \n [{}]'.format(s, u)
            labels.append(label)
        return labels
613

614
615
    def plot_corner(self, figsize=(7, 7), add_prior=False, nstds=None,
                    label_offset=0.4, dpi=300, rc_context={},
616
                    tglitch_ratio=False, fig_and_axes=None, save_fig=True,
617
                    **kwargs):
618
619
620
621
622
623
624
625
626
        """ 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)
627
628
629
        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.
630
631
632
633
634
635
636
637
638
639
640
641
642
643
        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
644
645
646
647
648
        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
649
650
        **kwargs:
            Passed to corner.corner
651

652
653
654
655
        Returns
        -------
        fig, axes:
            The matplotlib figure and axes, only returned if save_fig = False
656
657

        """
658

659
660
661
662
        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
663
664
        if self.ndim < 2:
            with plt.rc_context(rc_context):
665
666
667
668
                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
669
670
671
672
673
674
675
                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

676
        with plt.rc_context(rc_context):
677
678
679
680
681
            if fig_and_axes is None:
                fig, axes = plt.subplots(self.ndim, self.ndim,
                                         figsize=figsize)
            else:
                fig, axes = fig_and_axes
682
683

            samples_plt = copy.copy(self.samples)
684
            labels = self._get_labels()
685

686
            samples_plt = self._scale_samples(samples_plt, self.theta_keys)
687
688
689
690
691

            if tglitch_ratio:
                for j, k in enumerate(self.theta_keys):
                    if k == 'tglitch':
                        s = samples_plt[:, j]
692
693
694
                        samples_plt[:, j] = (
                            s - self.minStartTime)/(
                                self.maxStartTime - self.minStartTime)
695
                        labels[j] = r'$R_{\textrm{glitch}}$'
696
697
698
699
700
701
702

            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))
703
704
            elif 'range' in kwargs:
                _range = kwargs.pop('range')
705
706
707
            else:
                _range = None

708
709
710
711
            hist_kwargs = kwargs.pop('hist_kwargs', dict())
            if 'normed' not in hist_kwargs:
                hist_kwargs['normed'] = True

712
            fig_triangle = corner.corner(samples_plt,
713
                                         labels=labels,
714
715
716
717
718
719
720
721
722
                                         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,
723
                                         hist_kwargs=hist_kwargs,
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
                                         **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:
740
                self._add_prior_to_corner(axes, self.samples, add_prior)
741

742
743
744
745
746
            if save_fig:
                fig_triangle.savefig('{}/{}_corner.png'.format(
                    self.outdir, self.label), dpi=dpi)
            else:
                return fig, axes
747

748
    def _add_prior_to_corner(self, axes, samples, add_prior):
749
750
751
        for i, key in enumerate(self.theta_keys):
            ax = axes[i][i]
            s = samples[:, i]
752
753
754
755
756
757
758
759
760
761
            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)
762
763
            multiplier = self._get_rescale_multiplier_for_key(key)
            subtractor = self._get_rescale_subtractor_for_key(key)
764
765
766
767
768
769
770
771
            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])
772

773
774
775
776
777
778
779
780
    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]
781
            prior_func = self._generic_lnprior(**prior_dict)
782
783
784
785
786
            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
787
788
789
790
791
            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]
792
793
794
795
796
            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)
797
798
799
800
801
            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
802
803
804
805
806
            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]
807
808
809
            else:
                raise ValueError('Not implemented for prior type {}'.format(
                    prior_dict['type']))
810
            priorln = ax.plot(x, prior, 'C3', label='prior')
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
            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))

830
    def plot_cumulative_max(self, **kwargs):
831
832
833
834
        """ 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
835
836
837
838
        d, maxtwoF = self.get_max_twoF()
        for key, val in self.theta_prior.iteritems():
            if key not in d:
                d[key] = val
839

840
841
842
        if 'add_pfs' in kwargs:
            self.generate_loudest()

843
        if hasattr(self, 'search') is False:
844
            self._initiate_search_object()
845
846
847
        if self.binary is False:
            self.search.plot_twoF_cumulative(
                self.label, self.outdir, F0=d['F0'], F1=d['F1'], F2=d['F2'],
848
                Alpha=d['Alpha'], Delta=d['Delta'],
849
                tstart=self.minStartTime, tend=self.maxStartTime,
850
                **kwargs)
851
852
853
854
855
        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'],
856
                tstart=self.minStartTime, tend=self.maxStartTime, **kwargs)
Gregory Ashton's avatar
Gregory Ashton committed
857

858
    def _generic_lnprior(self, **kwargs):
859
860
861
862
        """ Return a lambda function of the pdf

        Parameters
        ----------
863
        **kwargs:
864
865
866
867
            A dictionary containing 'type' of pdf and shape parameters

        """

Gregory Ashton's avatar
Gregory Ashton committed
868
        def log_of_unif(x, a, b):
869
870
871
872
873
874
875
876
877
878
879
880
881
            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
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
        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):
898
            if x < loc:
899
900
901
902
903
904
905
906
907
908
909
910
911
912
                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
913
914
915
916
            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'])
917
        elif kwargs['type'] == 'halfnorm':
Gregory Ashton's avatar
Gregory Ashton committed
918
            return lambda x: log_of_halfnorm(x, kwargs['loc'], kwargs['scale'])
919
        elif kwargs['type'] == 'neghalfnorm':
Gregory Ashton's avatar
Gregory Ashton committed
920
921
            return lambda x: log_of_halfnorm(
                -x, kwargs['loc'], kwargs['scale'])
922
923
924
925
926
927
928
        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")

929
    def _generate_rv(self, **kwargs):
930
931
932
        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
933
934
935
        if dist_type == "log10unif":
            return 10**(np.random.uniform(low=kwargs['log10lower'],
                                          high=kwargs['log10upper']))
936
937
938
939
940
        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']))
941
942
943
        if dist_type == "neghalfnorm":
            return -1 * np.abs(np.random.normal(loc=kwargs['loc'],
                                                scale=kwargs['scale']))
944
945
946
947
948
949
        if dist_type == "lognorm":
            return np.random.lognormal(
                mean=kwargs['loc'], sigma=kwargs['scale'])
        else:
            raise ValueError("dist_type {} unknown".format(dist_type))

950
    def _plot_walkers(self, sampler, symbols=None, alpha=0.8, color="k",
951
952
                      temp=0, lw=0.1, nprod=0, add_det_stat_burnin=False,
                      fig=None, axes=None, xoffset=0, plot_det_stat=False,
953
                      context='ggplot', subtractions=None, labelpad=0.05):
954
955
        """ Plot all the chains from a sampler """

956
957
958
959
960
        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))

961
962
963
        if np.ndim(axes) > 1:
            axes = axes.flatten()

964
965
966
967
968
969
970
971
972
973
974
975
976
        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, :, :, :]

977
978
        if subtractions is None:
            subtractions = [0 for i in range(ndim)]
979
980
981
        else:
            if len(subtractions) != self.ndim:
                raise ValueError('subtractions must be of length ndim')
982

983
984
985
986
        if plot_det_stat:
            extra_subplots = 1
        else:
            extra_subplots = 0
987
        with plt.style.context((context)):
Gregory Ashton's avatar
Gregory Ashton committed
988
            plt.rcParams['text.usetex'] = True
Gregory Ashton's avatar
Gregory Ashton committed
989
            if fig is None and axes is None:
990
                fig = plt.figure(figsize=(4, 3.0*ndim))
991
992
                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
993
                               for i in range(2, ndim+1)]
994

Gregory Ashton's avatar
Gregory Ashton committed
995
            idxs = np.arange(chain.shape[1])
996
997
998
999
1000
            burnin_idx = chain.shape[1] - nprod
            if hasattr(self, 'convergence_idx'):
                convergence_idx = self.convergence_idx
            else:
                convergence_idx = burnin_idx