energy_operators.py 15.3 KB
Newer Older
Martin Reinecke's avatar
Martin Reinecke committed
1
2
3
4
5
6
7
8
9
10
11
12
13
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program.  If not, see <http://www.gnu.org/licenses/>.
#
14
# Copyright(C) 2013-2019 Max-Planck-Society
Martin Reinecke's avatar
Martin Reinecke committed
15
#
16
# NIFTy is being developed at the Max-Planck-Institut fuer Astrophysik.
Martin Reinecke's avatar
Martin Reinecke committed
17

Philipp Arras's avatar
Philipp Arras committed
18
19
import numpy as np

Philipp Arras's avatar
Philipp Arras committed
20
from .. import utilities
Martin Reinecke's avatar
Martin Reinecke committed
21
from ..domain_tuple import DomainTuple
Philipp Arras's avatar
Philipp Arras committed
22
23
from ..field import Field
from ..linearization import Linearization
Philipp Arras's avatar
Philipp Arras committed
24
25
26
from ..multi_domain import MultiDomain
from ..multi_field import MultiField
from ..sugar import makeDomain, makeOp
Philipp Arras's avatar
Philipp Arras committed
27
from .linear_operator import LinearOperator
Martin Reinecke's avatar
Martin Reinecke committed
28
from .operator import Operator
Martin Reinecke's avatar
fix    
Martin Reinecke committed
29
from .sampling_enabler import SamplingEnabler
Philipp Arras's avatar
Philipp Arras committed
30
from .sandwich_operator import SandwichOperator
31
from .scaling_operator import ScalingOperator
Philipp Arras's avatar
Philipp Arras committed
32
from .simple_linear_operators import FieldAdapter, VdotOperator
Martin Reinecke's avatar
Martin Reinecke committed
33
34
35


class EnergyOperator(Operator):
Philipp Arras's avatar
Philipp Arras committed
36
    """Operator which has a scalar domain as target domain.
37

Martin Reinecke's avatar
Martin Reinecke committed
38
    It is intended as an objective function for field inference.
39

Philipp Arras's avatar
Philipp Arras committed
40
41
42
    Examples
    --------
     - Information Hamiltonian, i.e. negative-log-probabilities.
Martin Reinecke's avatar
Martin Reinecke committed
43
     - Gibbs free energy, i.e. an averaged Hamiltonian, aka Kullback-Leibler
Philipp Arras's avatar
Philipp Arras committed
44
       divergence.
45
    """
Martin Reinecke's avatar
Martin Reinecke committed
46
47
48
    _target = DomainTuple.scalar_domain()


49
50
class Squared2NormOperator(EnergyOperator):
    """Computes the square of the L2-norm of the output of an operator.
51

Philipp Arras's avatar
Philipp Arras committed
52
53
54
    Parameters
    ----------
    domain : Domain, DomainTuple or tuple of Domain
55
        Domain of the operator in which the L2-norm shall be computed.
Martin Reinecke's avatar
Martin Reinecke committed
56
    """
Philipp Arras's avatar
Philipp Arras committed
57

Martin Reinecke's avatar
Martin Reinecke committed
58
59
60
61
    def __init__(self, domain):
        self._domain = domain

    def apply(self, x):
62
        self._check_input(x)
Martin Reinecke's avatar
Martin Reinecke committed
63
        if isinstance(x, Linearization):
Martin Reinecke's avatar
Martin Reinecke committed
64
            val = Field.scalar(x.val.vdot(x.val))
Martin Reinecke's avatar
Martin Reinecke committed
65
            jac = VdotOperator(2*x.val)(x.jac)
66
            return x.new(val, jac)
Martin Reinecke's avatar
Martin Reinecke committed
67
        return Field.scalar(x.vdot(x))
Martin Reinecke's avatar
Martin Reinecke committed
68

Martin Reinecke's avatar
Martin Reinecke committed
69

Martin Reinecke's avatar
Martin Reinecke committed
70
class QuadraticFormOperator(EnergyOperator):
Philipp Arras's avatar
Philipp Arras committed
71
    """Computes the L2-norm of a Field or MultiField with respect to a
72
    specific kernel given by `endo`.
Philipp Arras's avatar
Philipp Arras committed
73
74
75

    .. math ::
        E(f) = \\frac12 f^\\dagger \\text{endo}(f)
76
77
78

    Parameters
    ----------
Philipp Arras's avatar
Philipp Arras committed
79
    endo : EndomorphicOperator
80
         Kernel of the quadratic form
Martin Reinecke's avatar
Martin Reinecke committed
81
    """
Philipp Arras's avatar
Philipp Arras committed
82
83

    def __init__(self, endo):
Martin Reinecke's avatar
Martin Reinecke committed
84
        from .endomorphic_operator import EndomorphicOperator
Philipp Arras's avatar
Philipp Arras committed
85
        if not isinstance(endo, EndomorphicOperator):
Martin Reinecke's avatar
Martin Reinecke committed
86
            raise TypeError("op must be an EndomorphicOperator")
Philipp Arras's avatar
Philipp Arras committed
87
88
        self._op = endo
        self._domain = endo.domain
Martin Reinecke's avatar
Martin Reinecke committed
89
90

    def apply(self, x):
91
        self._check_input(x)
Martin Reinecke's avatar
Martin Reinecke committed
92
        if isinstance(x, Linearization):
Martin Reinecke's avatar
Martin Reinecke committed
93
94
            t1 = self._op(x.val)
            jac = VdotOperator(t1)(x.jac)
Martin Reinecke's avatar
Martin Reinecke committed
95
            val = Field.scalar(0.5*x.val.vdot(t1))
96
            return x.new(val, jac)
Martin Reinecke's avatar
Martin Reinecke committed
97
        return Field.scalar(0.5*x.vdot(self._op(x)))
Martin Reinecke's avatar
Martin Reinecke committed
98

Philipp Arras's avatar
Philipp Arras committed
99

100
class VariableCovarianceGaussianEnergy(EnergyOperator):
Reimar Leike's avatar
Reimar Leike committed
101
    """Computes the negative log pdf of a Gaussian with unknown covariance.
102

Reimar Leike's avatar
Reimar Leike committed
103
    The covariance is assumed to be diagonal.
104
105

    .. math ::
Reimar Leike's avatar
Reimar Leike committed
106
        E(s,D) = - \\log G(s, D) = 0.5 (s)^\\dagger D^{-1} (s) + 0.5 tr log(D),
107
108

    an information energy for a Gaussian distribution with residual s and
109
    diagonal covariance D.
Reimar Leike's avatar
Reimar Leike committed
110
111
    The domain of this energy will be a MultiDomain with two keys,
    the target will be the scalar domain.
112
113
114

    Parameters
    ----------
115
    domain : Domain, DomainTuple, tuple of Domain
Reimar Leike's avatar
Reimar Leike committed
116
        domain of the residual and domain of the covariance diagonal.
117

118
    residual : key
Philipp Arras's avatar
Philipp Arras committed
119
        Residual key of the Gaussian.
120

Philipp Arras's avatar
Philipp Arras committed
121
    inverse_covariance : key
122
        Inverse covariance diagonal key of the Gaussian.
123
124
    """

Philipp Arras's avatar
Philipp Arras committed
125
126
127
128
129
    def __init__(self, domain, residual_key, inverse_covariance_key):
        self._r = str(residual_key)
        self._icov = str(inverse_covariance_key)
        dom = DomainTuple.make(domain)
        self._domain = MultiDomain.make({self._r: dom, self._icov: dom})
130
131
132

    def apply(self, x):
        self._check_input(x)
133
134
135
        lin = isinstance(x, Linearization)
        r = FieldAdapter(self._domain[self._r], self._r)
        icov = FieldAdapter(self._domain[self._icov], self._icov)
136
137
        res0 = r.vdot(r*icov).real
        res1 = icov.log().sum()
Philipp Arras's avatar
Philipp Arras committed
138
        res = (res0-res1).scale(0.5)(x)
Philipp Arras's avatar
Philipp Arras committed
139
        if not lin or not x.want_metric:
Philipp Arras's avatar
Philipp Arras committed
140
            return res
Philipp Arras's avatar
Philipp Arras committed
141
142
        mf = {self._r: x.val[self._icov], self._icov: .5*x.val[self._icov]**(-2)}
        metric = makeOp(MultiField.from_dict(mf))
Philipp Arras's avatar
Fixup    
Philipp Arras committed
143
        return res.add_metric(SandwichOperator.make(x.jac, metric))
144

Martin Reinecke's avatar
Martin Reinecke committed
145
146

class GaussianEnergy(EnergyOperator):
Philipp Arras's avatar
Docs    
Philipp Arras committed
147
    """Computes a negative-log Gaussian.
148

Philipp Arras's avatar
Philipp Arras committed
149
    Represents up to constants in :math:`m`:
Martin Reinecke's avatar
Martin Reinecke committed
150

Philipp Arras's avatar
Philipp Arras committed
151
152
    .. math ::
        E(f) = - \\log G(f-m, D) = 0.5 (f-m)^\\dagger D^{-1} (f-m),
Martin Reinecke's avatar
cleanup    
Martin Reinecke committed
153

Philipp Arras's avatar
Philipp Arras committed
154
155
    an information energy for a Gaussian distribution with mean m and
    covariance D.
156

Philipp Arras's avatar
Philipp Arras committed
157
158
159
160
    Parameters
    ----------
    mean : Field
        Mean of the Gaussian. Default is 0.
161
162
    inverse_covariance : LinearOperator
        Inverse covariance of the Gaussian. Default is the identity operator.
Philipp Arras's avatar
Fixup    
Philipp Arras committed
163
    domain : Domain, DomainTuple, tuple of Domain or MultiDomain
Philipp Arras's avatar
Philipp Arras committed
164
165
166
167
168
169
        Operator domain. By default it is inferred from `mean` or
        `covariance` if specified

    Note
    ----
    At least one of the arguments has to be provided.
Martin Reinecke's avatar
Martin Reinecke committed
170
    """
Martin Reinecke's avatar
Martin Reinecke committed
171

172
    def __init__(self, mean=None, inverse_covariance=None, domain=None):
Martin Reinecke's avatar
Martin Reinecke committed
173
174
        if mean is not None and not isinstance(mean, (Field, MultiField)):
            raise TypeError
175
        if inverse_covariance is not None and not isinstance(inverse_covariance, LinearOperator):
Philipp Arras's avatar
Philipp Arras committed
176
177
            raise TypeError

Martin Reinecke's avatar
Martin Reinecke committed
178
179
180
        self._domain = None
        if mean is not None:
            self._checkEquivalence(mean.domain)
181
182
        if inverse_covariance is not None:
            self._checkEquivalence(inverse_covariance.domain)
Martin Reinecke's avatar
Martin Reinecke committed
183
184
185
186
187
        if domain is not None:
            self._checkEquivalence(domain)
        if self._domain is None:
            raise ValueError("no domain given")
        self._mean = mean
188
        if inverse_covariance is None:
189
            self._op = Squared2NormOperator(self._domain).scale(0.5)
Martin Reinecke's avatar
Martin Reinecke committed
190
        else:
191
192
            self._op = QuadraticFormOperator(inverse_covariance)
        self._icov = None if inverse_covariance is None else inverse_covariance
Martin Reinecke's avatar
Martin Reinecke committed
193
194

    def _checkEquivalence(self, newdom):
Martin Reinecke's avatar
fix    
Martin Reinecke committed
195
        newdom = makeDomain(newdom)
Martin Reinecke's avatar
Martin Reinecke committed
196
        if self._domain is None:
Philipp Arras's avatar
Philipp Arras committed
197
            self._domain = newdom
Martin Reinecke's avatar
Martin Reinecke committed
198
        else:
Philipp Arras's avatar
Philipp Arras committed
199
            if self._domain != newdom:
Martin Reinecke's avatar
Martin Reinecke committed
200
201
202
                raise ValueError("domain mismatch")

    def apply(self, x):
203
        self._check_input(x)
Philipp Arras's avatar
Philipp Arras committed
204
        residual = x if self._mean is None else x - self._mean
Philipp Arras's avatar
Changes    
Philipp Arras committed
205
        res = self._op(residual).real
206
        if not isinstance(x, Linearization) or not x.want_metric:
Martin Reinecke's avatar
Martin Reinecke committed
207
208
209
210
211
212
            return res
        metric = SandwichOperator.make(x.jac, self._icov)
        return res.add_metric(metric)


class PoissonianEnergy(EnergyOperator):
Philipp Arras's avatar
Docs    
Philipp Arras committed
213
214
    """Computes likelihood Hamiltonians of expected count field constrained by
    Poissonian count data.
215

Philipp Arras's avatar
Philipp Arras committed
216
    Represents up to an f-independent term :math:`log(d!)`:
217

Philipp Arras's avatar
Philipp Arras committed
218
219
    .. math ::
        E(f) = -\\log \\text{Poisson}(d|f) = \\sum f - d^\\dagger \\log(f),
220

Philipp Arras's avatar
Philipp Arras committed
221
    where f is a :class:`Field` in data space with the expectation values for
Martin Reinecke's avatar
Martin Reinecke committed
222
    the counts.
Philipp Arras's avatar
Philipp Arras committed
223
224
225
226
227
228

    Parameters
    ----------
    d : Field
        Data field with counts. Needs to have integer dtype and all field
        values need to be non-negative.
Martin Reinecke's avatar
Martin Reinecke committed
229
    """
Philipp Arras's avatar
Philipp Arras committed
230

231
    def __init__(self, d):
Philipp Arras's avatar
Philipp Arras committed
232
233
        if not isinstance(d, Field) or not np.issubdtype(d.dtype, np.integer):
            raise TypeError
Martin Reinecke's avatar
stage2    
Martin Reinecke committed
234
        if np.any(d.val < 0):
Philipp Arras's avatar
Philipp Arras committed
235
            raise ValueError
236
237
        self._d = d
        self._domain = DomainTuple.make(d.domain)
Martin Reinecke's avatar
Martin Reinecke committed
238
239

    def apply(self, x):
240
        self._check_input(x)
Philipp Arras's avatar
Philipp Arras committed
241
242
        fa = FieldAdapter(self._domain, 'foo')
        res = (fa.sum() - fa.log().vdot(self._d))(fa.adjoint(x))
Philipp Arras's avatar
Philipp Arras committed
243
        if not isinstance(x, Linearization) or not x.want_metric:
244
            return res
Martin Reinecke's avatar
Martin Reinecke committed
245
246
247
        metric = SandwichOperator.make(x.jac, makeOp(1./x.val))
        return res.add_metric(metric)

248

249
class InverseGammaLikelihood(EnergyOperator):
Philipp Arras's avatar
Docs    
Philipp Arras committed
250
    """Computes the negative log-likelihood of the inverse gamma distribution.
251
252
253

    It negative log-pdf(x) is given by

Martin Reinecke's avatar
Martin Reinecke committed
254
255
256
257
258
259
260
    .. math ::

        \\sum_i (\\alpha_i+1)*\\ln(x_i) + \\beta_i/x_i

    This is the likelihood for the variance :math:`x=S_k` given data
    :math:`\\beta = 0.5 |s_k|^2` where the Field :math:`s` is known to have
    the covariance :math:`S_k`.
261
262
263
264
265
266
267

    Parameters
    ----------
    beta : Field
        beta parameter of the inverse gamma distribution
    alpha : Scalar, Field, optional
        alpha parameter of the inverse gamma distribution
268
    """
Philipp Arras's avatar
Philipp Arras committed
269

270
271
    def __init__(self, beta, alpha=-0.5):
        if not isinstance(beta, Field):
Philipp Arras's avatar
Philipp Arras committed
272
            raise TypeError
Philipp Arras's avatar
Philipp Arras committed
273
        self._domain = DomainTuple.make(beta.domain)
274
275
        self._beta = beta
        if np.isscalar(alpha):
Martin Reinecke's avatar
stage2    
Martin Reinecke committed
276
            alpha = Field(beta.domain, np.full(beta.shape, alpha))
277
278
279
        elif not isinstance(alpha, Field):
            raise TypeError
        self._alphap1 = alpha+1
280
281

    def apply(self, x):
282
        self._check_input(x)
Philipp Arras's avatar
Philipp Arras committed
283
284
285
        fa = FieldAdapter(self._domain, 'foo')
        x = fa.adjoint(x)
        res = (fa.log().vdot(self._alphap1) + fa.one_over().vdot(self._beta))(x)
Philipp Arras's avatar
Philipp Arras committed
286
        if not isinstance(x, Linearization) or not x.want_metric:
287
            return res
288
        metric = SandwichOperator.make(x.jac, makeOp(self._alphap1/(x.val**2)))
289
290
291
        return res.add_metric(metric)


292
class StudentTEnergy(EnergyOperator):
Lukas Platz's avatar
Lukas Platz committed
293
    """Computes likelihood energy corresponding to Student's t-distribution.
294
295

    .. math ::
Lukas Platz's avatar
Lukas Platz committed
296
297
        E_\\theta(f) = -\\log \\text{StudentT}_\\theta(f)
                     = \\frac{\\theta + 1}{2} \\log(1 + \\frac{f^2}{\\theta}),
298

Lukas Platz's avatar
Lukas Platz committed
299
    where f is a field defined on `domain`.
300
301
302

    Parameters
    ----------
Lukas Platz's avatar
Lukas Platz committed
303
304
    domain : `Domain` or `DomainTuple`
        Domain of the operator
305
306
307
308
309
310
311
312
313
314
    theta : Scalar
        Degree of freedom parameter for the student t distribution
    """

    def __init__(self, domain, theta):
        self._domain = DomainTuple.make(domain)
        self._theta = theta

    def apply(self, x):
        self._check_input(x)
315
        v = ((self._theta+1)/2)*(x**2/self._theta).log1p().sum()
316
317
318
319
        if not isinstance(x, Linearization):
            return Field.scalar(v)
        if not x.want_metric:
            return v
320
        met = ScalingOperator(self.domain, (self._theta+1) / (self._theta+3))
321
322
323
324
        met = SandwichOperator.make(x.jac, met)
        return v.add_metric(met)


Martin Reinecke's avatar
Martin Reinecke committed
325
class BernoulliEnergy(EnergyOperator):
Philipp Arras's avatar
Philipp Arras committed
326
    """Computes likelihood energy of expected event frequency constrained by
327
328
    event data.

Philipp Arras's avatar
Philipp Arras committed
329
330
331
332
333
334
335
    .. math ::
        E(f) = -\\log \\text{Bernoulli}(d|f)
             = -d^\\dagger \\log f  - (1-d)^\\dagger \\log(1-f),

    where f is a field defined on `d.domain` with the expected
    frequencies of events.

336
337
    Parameters
    ----------
Martin Reinecke's avatar
Martin Reinecke committed
338
    d : Field
Philipp Arras's avatar
Philipp Arras committed
339
        Data field with events (1) or non-events (0).
Martin Reinecke's avatar
Martin Reinecke committed
340
    """
Philipp Arras's avatar
Philipp Arras committed
341

342
    def __init__(self, d):
Philipp Arras's avatar
Philipp Arras committed
343
344
        if not isinstance(d, Field) or not np.issubdtype(d.dtype, np.integer):
            raise TypeError
Martin Reinecke's avatar
stage2    
Martin Reinecke committed
345
        if not np.all(np.logical_or(d.val == 0, d.val == 1)):
Philipp Arras's avatar
Philipp Arras committed
346
            raise ValueError
Martin Reinecke's avatar
Martin Reinecke committed
347
        self._d = d
348
        self._domain = DomainTuple.make(d.domain)
Martin Reinecke's avatar
Martin Reinecke committed
349
350

    def apply(self, x):
351
        self._check_input(x)
Philipp Arras's avatar
Philipp Arras committed
352
353
        iden = FieldAdapter(self._domain, 'foo')
        from .adder import Adder
Philipp Arras's avatar
Philipp Arras committed
354
        v = -iden.log().vdot(self._d) + (Adder(1, domain=self._domain) @ iden.scale(-1)).log().vdot(self._d-1.)
Philipp Arras's avatar
Philipp Arras committed
355
        v = v(iden.adjoint(x))
Philipp Arras's avatar
Philipp Arras committed
356
        if not isinstance(x, Linearization) or not x.want_metric:
357
            return v
Philipp Arras's avatar
Philipp Arras committed
358
        met = makeOp(1./(x.val*(1. - x.val)))
Martin Reinecke's avatar
Martin Reinecke committed
359
360
361
362
        met = SandwichOperator.make(x.jac, met)
        return v.add_metric(met)


363
class StandardHamiltonian(EnergyOperator):
Philipp Arras's avatar
Philipp Arras committed
364
365
    """Computes an information Hamiltonian in its standard form, i.e. with the
    prior being a Gaussian with unit covariance.
366

Philipp Arras's avatar
Philipp Arras committed
367
    Let the likelihood energy be :math:`E_{lh}`. Then this operator computes:
368

Philipp Arras's avatar
Philipp Arras committed
369
370
    .. math ::
         H(f) = 0.5 f^\\dagger f + E_{lh}(f):
371

Martin Reinecke's avatar
Martin Reinecke committed
372
    Other field priors can be represented via transformations of a white
373
374
    Gaussian field into a field with the desired prior probability structure.

Martin Reinecke's avatar
Martin Reinecke committed
375
    By implementing prior information this way, the field prior is represented
376
377
378
    by a generative model, from which NIFTy can draw samples and infer a field
    using the Maximum a Posteriori (MAP) or the Variational Bayes (VB) method.

Philipp Arras's avatar
Philipp Arras committed
379
380
381
382
383
384
385
386
    The metric of this operator can be used as covariance for drawing Gaussian
    samples.

    Parameters
    ----------
    lh : EnergyOperator
        The likelihood energy.
    ic_samp : IterationController
387
        Tells an internal :class:`SamplingEnabler` which convergence criterion
Philipp Arras's avatar
Philipp Arras committed
388
389
390
391
392
393
        to use to draw Gaussian samples.

    See also
    --------
    `Encoding prior knowledge in the structure of the likelihood`,
    Jakob Knollmüller, Torsten A. Ensslin,
Martin Reinecke's avatar
Martin Reinecke committed
394
    `<https://arxiv.org/abs/1812.04403>`_
Martin Reinecke's avatar
Martin Reinecke committed
395
    """
Philipp Arras's avatar
Philipp Arras committed
396

397
    def __init__(self, lh, ic_samp=None, _c_inp=None):
Martin Reinecke's avatar
Martin Reinecke committed
398
399
        self._lh = lh
        self._prior = GaussianEnergy(domain=lh.domain)
400
401
        if _c_inp is not None:
            _, self._prior = self._prior.simplify_for_constant_input(_c_inp)
Martin Reinecke's avatar
Martin Reinecke committed
402
        self._ic_samp = ic_samp
Martin Reinecke's avatar
Martin Reinecke committed
403
        self._domain = lh.domain
Martin Reinecke's avatar
Martin Reinecke committed
404
405

    def apply(self, x):
406
        self._check_input(x)
Philipp Arras's avatar
Philipp Arras committed
407
408
        if (self._ic_samp is None or not isinstance(x, Linearization) or not x.want_metric):
            return (self._lh + self._prior)(x)
Martin Reinecke's avatar
Martin Reinecke committed
409
        else:
410
            lhx, prx = self._lh(x), self._prior(x)
Philipp Arras's avatar
Philipp Arras committed
411
            mtr = SamplingEnabler(lhx.metric, prx.metric, self._ic_samp)
Philipp Arras's avatar
Philipp Arras committed
412
            return (lhx + prx).add_metric(mtr)
Martin Reinecke's avatar
Martin Reinecke committed
413

Philipp Arras's avatar
Philipp Arras committed
414
415
    def __repr__(self):
        subs = 'Likelihood:\n{}'.format(utilities.indent(self._lh.__repr__()))
416
        subs += '\nPrior:\n{}'.format(self._prior)
Martin Reinecke's avatar
Martin Reinecke committed
417
        return 'StandardHamiltonian:\n' + utilities.indent(subs)
Philipp Arras's avatar
Philipp Arras committed
418

419
420
421
422
    def _simplify_for_constant_input_nontrivial(self, c_inp):
        out, lh1 = self._lh.simplify_for_constant_input(c_inp)
        return out, StandardHamiltonian(lh1, self._ic_samp, _c_inp=c_inp)

Martin Reinecke's avatar
Martin Reinecke committed
423

Martin Reinecke's avatar
Martin Reinecke committed
424
class AveragedEnergy(EnergyOperator):
Philipp Arras's avatar
Docs    
Philipp Arras committed
425
    """Averages an energy over samples.
Martin Reinecke's avatar
Martin Reinecke committed
426

427
428
429
    Parameters
    ----------
    h: Hamiltonian
Philipp Arras's avatar
Philipp Arras committed
430
       The energy to be averaged.
Martin Reinecke's avatar
Martin Reinecke committed
431
    res_samples : iterable of Fields
Torsten Ensslin's avatar
Torsten Ensslin committed
432
433
       Set of residual sample points to be added to mean field for
       approximate estimation of the KL.
434

Philipp Arras's avatar
Docs    
Philipp Arras committed
435
436
437
438
439
    Notes
    -----
    - Having symmetrized residual samples, with both :math:`v_i` and
      :math:`-v_i` being present, ensures that the distribution mean is
      exactly represented.
Torsten Ensslin's avatar
Fix te    
Torsten Ensslin committed
440

Philipp Arras's avatar
Docs    
Philipp Arras committed
441
442
443
    - :class:`AveragedEnergy(h)` approximates
      :math:`\\left< H(f) \\right>_{G(f-m,D)}` if the residuals :math:`f-m`
      are drawn from a Gaussian distribution with covariance :math:`D`.
Martin Reinecke's avatar
Martin Reinecke committed
444
    """
Martin Reinecke's avatar
Martin Reinecke committed
445
446
447

    def __init__(self, h, res_samples):
        self._h = h
Martin Reinecke's avatar
Martin Reinecke committed
448
        self._domain = h.domain
Martin Reinecke's avatar
Martin Reinecke committed
449
450
451
        self._res_samples = tuple(res_samples)

    def apply(self, x):
452
        self._check_input(x)
Philipp Arras's avatar
Philipp Arras committed
453
454
455
456
457
458
459
460
        if isinstance(self._domain, MultiDomain):
            iden = ScalingOperator(self._domain, 1.)
        else:
            iden = FieldAdapter(self._domain, 'foo')
            x = iden.adjoint(x)
        from .adder import Adder
        mymap = map(lambda v: self._h(Adder(v) @ iden), self._res_samples)
        return utilities.my_sum(mymap).scale(1./len(self._res_samples))(x)