pipeline.py 8.94 KB
Newer Older
1 2
# -*- coding: utf-8 -*-

3 4 5
import os
import numpy as np

6 7
from scipy import optimize

8 9
from mpi4py import MPI

10 11
from keepers import Loggable

12 13 14 15
from imagine.likelihoods import Likelihood
from imagine.magnetic_fields import MagneticFieldFactory
from imagine.observers import Observer
from imagine.priors import Prior
16
from imagine import pymultinest
17
from imagine.sample import Sample
18 19 20 21

comm = MPI.COMM_WORLD
size = comm.size
rank = comm.rank
22

23 24 25
WORK_TAG = 0
DIE_TAG = 1

26 27

class Pipeline(Loggable, object):
28 29 30 31 32 33 34 35 36 37 38 39
    """
    The pipeline
    - posses all the building blocks: magnetic_field, observer,
        likelihood and prior.
    - if multiple log-likelihoods and log-priors are given: sum the result
    - coordinates the repeated observation in order to compute an ensemble
    - controls which parameters of the magnetic field are tested
        (active parameters)


    """
    def __init__(self, magnetic_field_factory, observer, likelihood, prior,
40
                 active_variables=[], ensemble_size=1,
41
                 pymultinest_parameters={}, sample_callback=None):
42
        self.logger.debug("Setting up pipeline.")
43
        self.magnetic_field_factory = magnetic_field_factory
44 45 46
        self.observer = observer
        self.likelihood = likelihood
        self.prior = prior
47
        self.active_variables = active_variables
48
        self.ensemble_size = ensemble_size
49
        self.likelihood_rescaler = 1.
50

51
        # setting defaults for pymultinest
52 53 54 55 56
        self.pymultinest_parameters = {'verbose': True,
                                       'n_iter_before_update': 1,
                                       'n_live_points': 100}
        self.pymultinest_parameters.update(pymultinest_parameters)

57 58
        self.sample_callback = sample_callback

59 60
        self.fixed_random_seed = None

61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77
    @property
    def observer(self):
        return self._observer

    @observer.setter
    def observer(self, observer):
        if not isinstance(observer, Observer):
            raise TypeError("observer must be an instance of Observer-class.")
        self.logger.debug("Setting observer.")
        self._observer = observer

    @property
    def likelihood(self):
        return self._likelihood

    @likelihood.setter
    def likelihood(self, likelihood):
78
        self.logger.debug("Setting likelihood.")
79
        self._likelihood = ()
Theo Steininger's avatar
Theo Steininger committed
80
        if not (isinstance(likelihood, list) or
81 82
                isinstance(likelihood, tuple)):
            likelihood = [likelihood]
83 84 85 86 87
        for l in likelihood:
            if not isinstance(l, Likelihood):
                raise TypeError(
                    "likelihood must be an instance of Likelihood-class.")
            self._likelihood += (l,)
88 89 90 91 92 93 94

    @property
    def prior(self):
        return self._prior

    @prior.setter
    def prior(self, prior):
95
        self.logger.debug("Setting prior.")
96 97 98 99
        if not isinstance(prior, Prior):
            raise TypeError(
                "prior must be an instance of Prior-class.")
        self._prior = prior
100 101

    @property
102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128
    def magnetic_field_factory(self):
        return self._magnetic_field_factory

    @magnetic_field_factory.setter
    def magnetic_field_factory(self, magnetic_field_factory):
        if not isinstance(magnetic_field_factory, MagneticFieldFactory):
            raise TypeError(
                "magnetic_field_factory must be an instance of the "
                "MagneticFieldFactory-class.")
        self.logger.debug("Setting magnetic_field_factory.")
        self._magnetic_field_factory = magnetic_field_factory

    @property
    def active_variables(self):
        return self._active_variables

    @active_variables.setter
    def active_variables(self, active_variables):
        if not isinstance(active_variables, list):
            raise TypeError(
                    "active_variables must be a list.")
        self.logger.debug("Resetting active_variables to %s" %
                          str(active_variables))
        new_active = []
        for av in active_variables:
            new_active += [str(av)]
        self._active_variables = new_active
129 130 131 132 133 134 135 136 137 138 139 140 141 142

    @property
    def ensemble_size(self):
        return self._ensemble_size

    @ensemble_size.setter
    def ensemble_size(self, ensemble_size):

        ensemble_size = int(ensemble_size)
        if ensemble_size <= 0:
            raise ValueError("ensemble_size must be positive!")
        self.logger.debug("Setting ensemble size to %i." % ensemble_size)
        self._ensemble_size = ensemble_size

143 144 145 146 147 148 149 150
    def _multinest_likelihood(self, cube, ndim, nparams):
        cube_content = np.empty(ndim)
        for i in xrange(ndim):
            cube_content[i] = cube[i]
        if rank != 0:
            raise RuntimeError("_multinest_likelihood must only be called on "
                               "rank==0.")
        for i in xrange(1, size):
151
            comm.send(cube_content, dest=i, tag=WORK_TAG)
Theo Steininger's avatar
Theo Steininger committed
152
        self.logger.debug("Sent multinest-cube to nodes with rank > 0.")
153 154 155 156

        return self._core_likelihood(cube_content)

    def _listen_for_likelihood_calls(self):
157 158 159 160 161 162 163 164
        status = MPI.Status()
        while True:
            cube = comm.recv(source=0, tag=MPI.ANY_TAG, status=status)
            if status == DIE_TAG:
                self.logger.debug("Received DIE_TAG from rank 0.")
                break
            self.logger.debug("Received cube from rank 0.")
            self._core_likelihood(cube)
165 166

    def _core_likelihood(self, cube):
167 168
        self.logger.debug("Beginning Likelihood-calculation for %s." %
                          str(cube))
169 170 171 172 173 174
        # translate cube to variables
        variables = {}
        for i, av in enumerate(self.active_variables):
            variables[av] = cube[i]

        # create magnetic field
175 176
        self.logger.debug("Creating magnetic field.")
        b_field = self.magnetic_field_factory.generate(
177 178 179
                                          variables=variables,
                                          ensemble_size=self.ensemble_size,
                                          random_seed=self.fixed_random_seed)
180

181
        # create observables
182
        self.logger.debug("Creating observables.")
183 184 185
        observables = self.observer(b_field)

        # add up individual log-likelihood terms
186
        self.logger.debug("Evaluating likelihood(s).")
187 188
        likelihood = ()
        total_likelihood = 0
189
        for like in self.likelihood:
190 191 192
            current_likelihood = like(observables)
            likelihood += (current_likelihood, )
            total_likelihood += current_likelihood
193

Theo Steininger's avatar
Theo Steininger committed
194
        self.logger.info("Evaluated likelihood: %f for %s" %
195 196 197 198 199 200 201 202 203 204 205
                         (total_likelihood, str(cube)))

        if self.sample_callback is not None:
            self.logger.debug("Creating sample-object.")
            sample = Sample(variables=variables,
                            magnetic_field=b_field,
                            observables=observables,
                            likelihood=likelihood,
                            total_likelihood=total_likelihood)
            self.sample_callback(sample)

206
        return total_likelihood * self.likelihood_rescaler
207

208
    def __call__(self):
209 210 211

        if rank == 0:
            # kickstart pymultinest
Theo Steininger's avatar
Theo Steininger committed
212
            self.logger.info("Starting pymultinest.")
213 214 215 216 217
            if not os.path.exists("chains"):
                os.mkdir("chains")
            pymultinest.run(self._multinest_likelihood,
                            self.prior,
                            len(self.active_variables),
218
                            **self.pymultinest_parameters)
219 220 221 222
            self.logger.info("pymultinest finished.")
            for i in xrange(1, size):
                self.logger.debug("Sending DIE_TAG to rank %i." % i)
                comm.send(None, dest=i, tag=DIE_TAG)
223 224 225
        else:
            # let all other nodes listen for likelihood evaluations
            self._listen_for_likelihood_calls()
226

227 228
    def find_minimum(self, starting_guess=None, method='Nelder-Mead',
                     **kwargs):
229 230 231 232 233 234
        if starting_guess is None:
            starting_guess = np.zeros(len(self.active_variables)) + 0.5

        if rank == 0:
            # kickstart pymultinest
            self.logger.info("Starting minimizer.")
Theo Steininger's avatar
Theo Steininger committed
235
            call_func = lambda z: -self._multinest_likelihood(
236 237 238
                                                 z,
                                                 len(self.active_variables),
                                                 len(self.active_variables))
239 240 241 242
            minimum = optimize.minimize(func=call_func,
                                        x0=starting_guess,
                                        method=method,
                                        **kwargs)
243 244 245 246 247 248 249 250 251 252
            self.logger.info("Minimizer finished.")
            for i in xrange(1, size):
                self.logger.debug("Sending DIE_TAG to rank %i." % i)
                comm.send(None, dest=i, tag=DIE_TAG)
        else:
            minimum = None
            # let all other nodes listen for likelihood evaluations
            self._listen_for_likelihood_calls()
        minimum = comm.bcast(minimum, root=0)
        return minimum