energy_operators.py 16.4 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/>.
#
Philipp Arras's avatar
Philipp Arras committed
14
# Copyright(C) 2013-2020 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
from ..field import Field
Philipp Arras's avatar
Philipp Arras committed
23
24
25
from ..multi_domain import MultiDomain
from ..multi_field import MultiField
from ..sugar import makeDomain, makeOp
Philipp Arras's avatar
Philipp Arras committed
26
from .linear_operator import LinearOperator
Martin Reinecke's avatar
Martin Reinecke committed
27
from .operator import Operator
28
from .sampling_enabler import SamplingDtypeSetter, SamplingEnabler
29
from .scaling_operator import ScalingOperator
Philipp Arras's avatar
Cleanup    
Philipp Arras committed
30
from .simple_linear_operators import VdotOperator
Philipp Arras's avatar
Philipp Arras committed
31
32
33
34
35
36


def _check_sampling_dtype(domain, dtypes):
    if dtypes is None:
        return
    if isinstance(domain, DomainTuple):
Philipp Arras's avatar
Philipp Arras committed
37
38
        np.dtype(dtypes)
        return
Philipp Arras's avatar
Philipp Arras committed
39
    elif isinstance(domain, MultiDomain):
Philipp Arras's avatar
Philipp Arras committed
40
41
42
43
44
45
46
        if isinstance(dtypes, dict):
            for dt in dtypes.values():
                np.dtype(dt)
            if set(domain.keys()) == set(dtypes.keys()):
                return
        else:
            np.dtype(dtypes)
Philipp Arras's avatar
Philipp Arras committed
47
            return
Philipp Arras's avatar
Philipp Arras committed
48
    raise TypeError
Philipp Arras's avatar
Philipp Arras committed
49
50
51
52
53
54
55
56
57
58
59
60
61


def _field_to_dtype(field):
    if isinstance(field, Field):
        dt = field.dtype
    elif isinstance(field, MultiField):
        dt = {kk: ff.dtype for kk, ff in field.items()}
    else:
        raise TypeError
    _check_sampling_dtype(field.domain, dt)
    return dt


Martin Reinecke's avatar
Martin Reinecke committed
62
class EnergyOperator(Operator):
Philipp Arras's avatar
Philipp Arras committed
63
    """Operator which has a scalar domain as target domain.
64

Martin Reinecke's avatar
Martin Reinecke committed
65
    It is intended as an objective function for field inference.
66

Philipp Arras's avatar
Philipp Arras committed
67
68
69
    Examples
    --------
     - Information Hamiltonian, i.e. negative-log-probabilities.
Martin Reinecke's avatar
Martin Reinecke committed
70
     - Gibbs free energy, i.e. an averaged Hamiltonian, aka Kullback-Leibler
Philipp Arras's avatar
Philipp Arras committed
71
       divergence.
72
    """
Martin Reinecke's avatar
Martin Reinecke committed
73
74
75
    _target = DomainTuple.scalar_domain()


76
77
class Squared2NormOperator(EnergyOperator):
    """Computes the square of the L2-norm of the output of an operator.
78

Philipp Arras's avatar
Philipp Arras committed
79
80
81
    Parameters
    ----------
    domain : Domain, DomainTuple or tuple of Domain
82
        Domain of the operator in which the L2-norm shall be computed.
Martin Reinecke's avatar
Martin Reinecke committed
83
    """
Philipp Arras's avatar
Philipp Arras committed
84

Martin Reinecke's avatar
Martin Reinecke committed
85
86
87
    def __init__(self, domain):
        self._domain = domain

Philipp Arras's avatar
Philipp Arras committed
88
    def apply(self, x):
89
        self._check_input(x)
Martin Reinecke's avatar
more    
Martin Reinecke committed
90
91
        if x.jac is None:
            return x.vdot(x)
Philipp Arras's avatar
Philipp Arras committed
92
93
        res = x.val.vdot(x.val)
        return x.new(res, VdotOperator(2*x.val))
Martin Reinecke's avatar
Martin Reinecke committed
94

Martin Reinecke's avatar
Martin Reinecke committed
95

Martin Reinecke's avatar
Martin Reinecke committed
96
class QuadraticFormOperator(EnergyOperator):
Philipp Arras's avatar
Philipp Arras committed
97
    """Computes the L2-norm of a Field or MultiField with respect to a
98
    specific kernel given by `endo`.
Philipp Arras's avatar
Philipp Arras committed
99
100
101

    .. math ::
        E(f) = \\frac12 f^\\dagger \\text{endo}(f)
102
103
104

    Parameters
    ----------
Philipp Arras's avatar
Philipp Arras committed
105
    endo : EndomorphicOperator
106
         Kernel of the quadratic form
Martin Reinecke's avatar
Martin Reinecke committed
107
    """
Philipp Arras's avatar
Philipp Arras committed
108
109

    def __init__(self, endo):
Martin Reinecke's avatar
Martin Reinecke committed
110
        from .endomorphic_operator import EndomorphicOperator
Philipp Arras's avatar
Philipp Arras committed
111
        if not isinstance(endo, EndomorphicOperator):
Martin Reinecke's avatar
Martin Reinecke committed
112
            raise TypeError("op must be an EndomorphicOperator")
Philipp Arras's avatar
Philipp Arras committed
113
114
        self._op = endo
        self._domain = endo.domain
Martin Reinecke's avatar
Martin Reinecke committed
115

Philipp Arras's avatar
Philipp Arras committed
116
    def apply(self, x):
117
        self._check_input(x)
Martin Reinecke's avatar
more    
Martin Reinecke committed
118
        if x.jac is None:
Philipp Arras's avatar
Philipp Arras committed
119
120
121
            return 0.5*x.vdot(self._op(x))
        res = 0.5*x.val.vdot(self._op(x.val))
        return x.new(res, VdotOperator(self._op(x.val)))
Martin Reinecke's avatar
Martin Reinecke committed
122

Philipp Arras's avatar
Philipp Arras committed
123

124
class VariableCovarianceGaussianEnergy(EnergyOperator):
Reimar Leike's avatar
Reimar Leike committed
125
    """Computes the negative log pdf of a Gaussian with unknown covariance.
126

Reimar Leike's avatar
Reimar Leike committed
127
    The covariance is assumed to be diagonal.
128
129

    .. math ::
Reimar Leike's avatar
Reimar Leike committed
130
        E(s,D) = - \\log G(s, D) = 0.5 (s)^\\dagger D^{-1} (s) + 0.5 tr log(D),
131
132

    an information energy for a Gaussian distribution with residual s and
133
    diagonal covariance D.
Reimar Leike's avatar
Reimar Leike committed
134
135
    The domain of this energy will be a MultiDomain with two keys,
    the target will be the scalar domain.
136
137
138

    Parameters
    ----------
139
    domain : Domain, DomainTuple, tuple of Domain
Reimar Leike's avatar
Reimar Leike committed
140
        domain of the residual and domain of the covariance diagonal.
141

142
    residual : key
Philipp Arras's avatar
Philipp Arras committed
143
        Residual key of the Gaussian.
144

Philipp Arras's avatar
Philipp Arras committed
145
    inverse_covariance : key
146
        Inverse covariance diagonal key of the Gaussian.
Philipp Arras's avatar
Philipp Arras committed
147

148
    sampling_dtype : np.dtype
Philipp Arras's avatar
Philipp Arras committed
149
        Data type of the samples. Usually either 'np.float*' or 'np.complex*'
150
151
    """

Philipp Arras's avatar
Philipp Arras committed
152
    def __init__(self, domain, residual_key, inverse_covariance_key, sampling_dtype):
Philipp Arras's avatar
Philipp Arras committed
153
154
        self._kr = str(residual_key)
        self._ki = str(inverse_covariance_key)
Philipp Arras's avatar
Philipp Arras committed
155
        dom = DomainTuple.make(domain)
Philipp Arras's avatar
Philipp Arras committed
156
        self._domain = MultiDomain.make({self._kr: dom, self._ki: dom})
Philipp Arras's avatar
Philipp Arras committed
157
158
159
        self._dt = {self._kr: sampling_dtype, self._ki: np.float64}
        _check_sampling_dtype(self._domain, self._dt)
        self._cplx = np.issubdtype(sampling_dtype, np.complexfloating)
160

Philipp Arras's avatar
Philipp Arras committed
161
    def apply(self, x):
162
        self._check_input(x)
Philipp Arras's avatar
Philipp Arras committed
163
        r, i = x[self._kr], x[self._ki]
Philipp Arras's avatar
Philipp Arras committed
164
165
166
167
        if self._cplx:
            res = 0.5*r.vdot(r*i.real).real - i.ptw("log").sum()
        else:
            res = 0.5*(r.vdot(r*i) - i.ptw("log").sum())
Martin Reinecke's avatar
more    
Martin Reinecke committed
168
        if not x.want_metric:
Philipp Arras's avatar
Philipp Arras committed
169
            return res
Philipp Arras's avatar
Philipp Arras committed
170
171
        met = i.val if self._cplx else 0.5*i.val
        met = MultiField.from_dict({self._kr: i.val, self._ki: met**(-2)})
Philipp Arras's avatar
Philipp Arras committed
172
        return res.add_metric(SamplingDtypeSetter(makeOp(met), self._dt))
173

Martin Reinecke's avatar
Martin Reinecke committed
174
175

class GaussianEnergy(EnergyOperator):
Philipp Arras's avatar
Docs    
Philipp Arras committed
176
    """Computes a negative-log Gaussian.
177

Philipp Arras's avatar
Philipp Arras committed
178
    Represents up to constants in :math:`m`:
Martin Reinecke's avatar
Martin Reinecke committed
179

Philipp Arras's avatar
Philipp Arras committed
180
181
    .. 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
182

Philipp Arras's avatar
Philipp Arras committed
183
184
    an information energy for a Gaussian distribution with mean m and
    covariance D.
185

Philipp Arras's avatar
Philipp Arras committed
186
187
188
189
    Parameters
    ----------
    mean : Field
        Mean of the Gaussian. Default is 0.
190
191
    inverse_covariance : LinearOperator
        Inverse covariance of the Gaussian. Default is the identity operator.
Philipp Arras's avatar
Fixup    
Philipp Arras committed
192
    domain : Domain, DomainTuple, tuple of Domain or MultiDomain
Philipp Arras's avatar
Philipp Arras committed
193
194
195
196
197
198
        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
199
    """
Martin Reinecke's avatar
Martin Reinecke committed
200

Philipp Arras's avatar
Philipp Arras committed
201
    def __init__(self, mean=None, inverse_covariance=None, domain=None, sampling_dtype=None):
Martin Reinecke's avatar
Martin Reinecke committed
202
203
        if mean is not None and not isinstance(mean, (Field, MultiField)):
            raise TypeError
204
        if inverse_covariance is not None and not isinstance(inverse_covariance, LinearOperator):
Philipp Arras's avatar
Philipp Arras committed
205
206
            raise TypeError

Martin Reinecke's avatar
Martin Reinecke committed
207
208
209
        self._domain = None
        if mean is not None:
            self._checkEquivalence(mean.domain)
210
211
        if inverse_covariance is not None:
            self._checkEquivalence(inverse_covariance.domain)
Martin Reinecke's avatar
Martin Reinecke committed
212
213
214
215
216
        if domain is not None:
            self._checkEquivalence(domain)
        if self._domain is None:
            raise ValueError("no domain given")
        self._mean = mean
Philipp Arras's avatar
Philipp Arras committed
217
218
219
220
221
222
223
224
225
226
227

        # Infer sampling dtype
        if self._mean is None:
            _check_sampling_dtype(self._domain, sampling_dtype)
        else:
            if sampling_dtype is None:
                sampling_dtype = _field_to_dtype(self._mean)
            else:
                if sampling_dtype != _field_to_dtype(self._mean):
                    raise ValueError("Sampling dtype and mean not compatible")

228
        if inverse_covariance is None:
229
            self._op = Squared2NormOperator(self._domain).scale(0.5)
Philipp Arras's avatar
Philipp Arras committed
230
            self._met = ScalingOperator(self._domain, 1)
231
            self._trivial_invcov = True
Martin Reinecke's avatar
Martin Reinecke committed
232
        else:
233
            self._op = QuadraticFormOperator(inverse_covariance)
Philipp Arras's avatar
Philipp Arras committed
234
            self._met = inverse_covariance
235
            self._trivial_invcov = False
Philipp Arras's avatar
Philipp Arras committed
236
        if sampling_dtype is not None:
237
            self._met = SamplingDtypeSetter(self._met, sampling_dtype)
Martin Reinecke's avatar
Martin Reinecke committed
238
239

    def _checkEquivalence(self, newdom):
Martin Reinecke's avatar
fix    
Martin Reinecke committed
240
        newdom = makeDomain(newdom)
Martin Reinecke's avatar
Martin Reinecke committed
241
        if self._domain is None:
Philipp Arras's avatar
Philipp Arras committed
242
            self._domain = newdom
Martin Reinecke's avatar
Martin Reinecke committed
243
        else:
Philipp Arras's avatar
Philipp Arras committed
244
            if self._domain != newdom:
Martin Reinecke's avatar
Martin Reinecke committed
245
246
                raise ValueError("domain mismatch")

Philipp Arras's avatar
Philipp Arras committed
247
    def apply(self, x):
248
        self._check_input(x)
Philipp Arras's avatar
Philipp Arras committed
249
        residual = x if self._mean is None else x - self._mean
Philipp Arras's avatar
Changes    
Philipp Arras committed
250
        res = self._op(residual).real
Martin Reinecke's avatar
more    
Martin Reinecke committed
251
        if x.want_metric:
Philipp Arras's avatar
Philipp Arras committed
252
253
            return res.add_metric(self._met)
        return res
Martin Reinecke's avatar
Martin Reinecke committed
254

Philipp Arras's avatar
Philipp Arras committed
255
256
257
258
    def __repr__(self):
        dom = '()' if isinstance(self.domain, DomainTuple) else self.domain.keys()
        return f'GaussianEnergy {dom}'

Martin Reinecke's avatar
Martin Reinecke committed
259
260

class PoissonianEnergy(EnergyOperator):
Philipp Arras's avatar
Docs    
Philipp Arras committed
261
262
    """Computes likelihood Hamiltonians of expected count field constrained by
    Poissonian count data.
263

Philipp Arras's avatar
Philipp Arras committed
264
    Represents up to an f-independent term :math:`log(d!)`:
265

Philipp Arras's avatar
Philipp Arras committed
266
267
    .. math ::
        E(f) = -\\log \\text{Poisson}(d|f) = \\sum f - d^\\dagger \\log(f),
268

Philipp Arras's avatar
Philipp Arras committed
269
    where f is a :class:`Field` in data space with the expectation values for
Martin Reinecke's avatar
Martin Reinecke committed
270
    the counts.
Philipp Arras's avatar
Philipp Arras committed
271
272
273
274
275
276

    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
277
    """
Philipp Arras's avatar
Philipp Arras committed
278

279
    def __init__(self, d):
Philipp Arras's avatar
Philipp Arras committed
280
281
        if not isinstance(d, Field) or not np.issubdtype(d.dtype, np.integer):
            raise TypeError
Martin Reinecke's avatar
stage2    
Martin Reinecke committed
282
        if np.any(d.val < 0):
Philipp Arras's avatar
Philipp Arras committed
283
            raise ValueError
284
285
        self._d = d
        self._domain = DomainTuple.make(d.domain)
Martin Reinecke's avatar
Martin Reinecke committed
286

Philipp Arras's avatar
Philipp Arras committed
287
    def apply(self, x):
288
        self._check_input(x)
Martin Reinecke's avatar
Martin Reinecke committed
289
        res = x.sum() - x.ptw("log").vdot(self._d)
Martin Reinecke's avatar
more    
Martin Reinecke committed
290
        if not x.want_metric:
291
            return res
292
        return res.add_metric(SamplingDtypeSetter(makeOp(1./x.val), np.float64))
Martin Reinecke's avatar
Martin Reinecke committed
293

294

295
class InverseGammaLikelihood(EnergyOperator):
Philipp Arras's avatar
Docs    
Philipp Arras committed
296
    """Computes the negative log-likelihood of the inverse gamma distribution.
297
298
299

    It negative log-pdf(x) is given by

Martin Reinecke's avatar
Martin Reinecke committed
300
301
302
303
304
305
306
    .. 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`.
307
308
309
310
311
312
313

    Parameters
    ----------
    beta : Field
        beta parameter of the inverse gamma distribution
    alpha : Scalar, Field, optional
        alpha parameter of the inverse gamma distribution
314
    """
Philipp Arras's avatar
Philipp Arras committed
315

316
317
    def __init__(self, beta, alpha=-0.5):
        if not isinstance(beta, Field):
Philipp Arras's avatar
Philipp Arras committed
318
            raise TypeError
Philipp Arras's avatar
Philipp Arras committed
319
        self._domain = DomainTuple.make(beta.domain)
320
321
        self._beta = beta
        if np.isscalar(alpha):
Martin Reinecke's avatar
stage2    
Martin Reinecke committed
322
            alpha = Field(beta.domain, np.full(beta.shape, alpha))
323
324
325
        elif not isinstance(alpha, Field):
            raise TypeError
        self._alphap1 = alpha+1
Philipp Arras's avatar
Philipp Arras committed
326
327
328
329
        if not self._beta.dtype == np.float64:
            # FIXME Add proper complex support for this energy
            raise TypeError
        self._sampling_dtype = _field_to_dtype(self._beta)
330

Philipp Arras's avatar
Philipp Arras committed
331
    def apply(self, x):
332
        self._check_input(x)
Martin Reinecke's avatar
Martin Reinecke committed
333
        res = x.ptw("log").vdot(self._alphap1) + x.ptw("reciprocal").vdot(self._beta)
Martin Reinecke's avatar
more    
Martin Reinecke committed
334
        if not x.want_metric:
335
            return res
Philipp Arras's avatar
Philipp Arras committed
336
337
        met = makeOp(self._alphap1/(x.val**2))
        if self._sampling_dtype is not None:
338
            met = SamplingDtypeSetter(met, self._sampling_dtype)
Philipp Arras's avatar
Philipp Arras committed
339
        return res.add_metric(met)
340
341


342
class StudentTEnergy(EnergyOperator):
Lukas Platz's avatar
Lukas Platz committed
343
    """Computes likelihood energy corresponding to Student's t-distribution.
344
345

    .. math ::
Lukas Platz's avatar
Lukas Platz committed
346
347
        E_\\theta(f) = -\\log \\text{StudentT}_\\theta(f)
                     = \\frac{\\theta + 1}{2} \\log(1 + \\frac{f^2}{\\theta}),
348

Philipp Arras's avatar
Philipp Arras committed
349
350
    where f is a field defined on `domain`. Assumes that the data is `float64`
    for sampling.
351
352
353

    Parameters
    ----------
Lukas Platz's avatar
Lukas Platz committed
354
355
    domain : `Domain` or `DomainTuple`
        Domain of the operator
Reimar Leike's avatar
Reimar Leike committed
356
    theta : Scalar or Field
357
358
359
        Degree of freedom parameter for the student t distribution
    """

Philipp Arras's avatar
Philipp Arras committed
360
    def __init__(self, domain, theta):
361
362
363
        self._domain = DomainTuple.make(domain)
        self._theta = theta

Philipp Arras's avatar
Philipp Arras committed
364
    def apply(self, x):
365
        self._check_input(x)
366
        res = (((self._theta+1)/2)*(x**2/self._theta).ptw("log1p")).sum()
Martin Reinecke's avatar
more    
Martin Reinecke committed
367
        if not x.want_metric:
Philipp Arras's avatar
Philipp Arras committed
368
            return res
369
        met = makeOp((self._theta+1) / (self._theta+3), self.domain)
Philipp Arras's avatar
Philipp Arras committed
370
        return res.add_metric(SamplingDtypeSetter(met, np.float64))
371
372


Martin Reinecke's avatar
Martin Reinecke committed
373
class BernoulliEnergy(EnergyOperator):
Philipp Arras's avatar
Philipp Arras committed
374
    """Computes likelihood energy of expected event frequency constrained by
375
376
    event data.

Philipp Arras's avatar
Philipp Arras committed
377
378
379
380
381
382
383
    .. 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.

384
385
    Parameters
    ----------
Martin Reinecke's avatar
Martin Reinecke committed
386
    d : Field
Philipp Arras's avatar
Philipp Arras committed
387
        Data field with events (1) or non-events (0).
Martin Reinecke's avatar
Martin Reinecke committed
388
    """
Philipp Arras's avatar
Philipp Arras committed
389

390
    def __init__(self, d):
Philipp Arras's avatar
Philipp Arras committed
391
392
        if not isinstance(d, Field) or not np.issubdtype(d.dtype, np.integer):
            raise TypeError
Philipp Arras's avatar
Philipp Arras committed
393
        if np.any(np.logical_and(d.val != 0, d.val != 1)):
Philipp Arras's avatar
Philipp Arras committed
394
            raise ValueError
Martin Reinecke's avatar
Martin Reinecke committed
395
        self._d = d
396
        self._domain = DomainTuple.make(d.domain)
Martin Reinecke's avatar
Martin Reinecke committed
397

Philipp Arras's avatar
Philipp Arras committed
398
    def apply(self, x):
399
        self._check_input(x)
Martin Reinecke's avatar
Martin Reinecke committed
400
        res = -x.ptw("log").vdot(self._d) + (1.-x).ptw("log").vdot(self._d-1.)
Martin Reinecke's avatar
more    
Martin Reinecke committed
401
        if not x.want_metric:
Philipp Arras's avatar
Philipp Arras committed
402
            return res
Philipp Arras's avatar
Philipp Arras committed
403
        met = makeOp(1./(x.val*(1. - x.val)))
404
        return res.add_metric(SamplingDtypeSetter(met, np.float64))
Martin Reinecke's avatar
Martin Reinecke committed
405
406


407
class StandardHamiltonian(EnergyOperator):
Philipp Arras's avatar
Philipp Arras committed
408
409
    """Computes an information Hamiltonian in its standard form, i.e. with the
    prior being a Gaussian with unit covariance.
410

Philipp Arras's avatar
Philipp Arras committed
411
    Let the likelihood energy be :math:`E_{lh}`. Then this operator computes:
412

Philipp Arras's avatar
Philipp Arras committed
413
414
    .. math ::
         H(f) = 0.5 f^\\dagger f + E_{lh}(f):
415

Martin Reinecke's avatar
Martin Reinecke committed
416
    Other field priors can be represented via transformations of a white
417
418
    Gaussian field into a field with the desired prior probability structure.

Martin Reinecke's avatar
Martin Reinecke committed
419
    By implementing prior information this way, the field prior is represented
420
421
422
    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
423
424
425
426
427
428
429
430
    The metric of this operator can be used as covariance for drawing Gaussian
    samples.

    Parameters
    ----------
    lh : EnergyOperator
        The likelihood energy.
    ic_samp : IterationController
431
        Tells an internal :class:`SamplingEnabler` which convergence criterion
Philipp Arras's avatar
Philipp Arras committed
432
        to use to draw Gaussian samples.
Philipp Arras's avatar
Philipp Arras committed
433
434
    prior_dtype : numpy.dtype or dict of numpy.dtype, optional
        Data type of prior used for sampling.
Philipp Arras's avatar
Philipp Arras committed
435
436
437
438
439

    See also
    --------
    `Encoding prior knowledge in the structure of the likelihood`,
    Jakob Knollmüller, Torsten A. Ensslin,
Martin Reinecke's avatar
Martin Reinecke committed
440
    `<https://arxiv.org/abs/1812.04403>`_
Martin Reinecke's avatar
Martin Reinecke committed
441
    """
Philipp Arras's avatar
Philipp Arras committed
442

Philipp Arras's avatar
Philipp Arras committed
443
    def __init__(self, lh, ic_samp=None, _c_inp=None, prior_dtype=np.float64):
Martin Reinecke's avatar
Martin Reinecke committed
444
        self._lh = lh
Philipp Arras's avatar
Philipp Arras committed
445
        self._prior = GaussianEnergy(domain=lh.domain, sampling_dtype=prior_dtype)
446
447
        if _c_inp is not None:
            _, self._prior = self._prior.simplify_for_constant_input(_c_inp)
Martin Reinecke's avatar
Martin Reinecke committed
448
        self._ic_samp = ic_samp
Martin Reinecke's avatar
Martin Reinecke committed
449
        self._domain = lh.domain
Martin Reinecke's avatar
Martin Reinecke committed
450

Philipp Arras's avatar
Philipp Arras committed
451
    def apply(self, x):
452
        self._check_input(x)
Martin Reinecke's avatar
more    
Martin Reinecke committed
453
        if not x.want_metric or self._ic_samp is None:
Philipp Arras's avatar
Philipp Arras committed
454
            return (self._lh + self._prior)(x)
Philipp Arras's avatar
Philipp Arras committed
455
456
        lhx, prx = self._lh(x), self._prior(x)
        return (lhx+prx).add_metric(SamplingEnabler(lhx.metric, prx.metric, self._ic_samp))
Martin Reinecke's avatar
Martin Reinecke committed
457

Philipp Arras's avatar
Philipp Arras committed
458
459
    def __repr__(self):
        subs = 'Likelihood:\n{}'.format(utilities.indent(self._lh.__repr__()))
460
        subs += '\nPrior:\n{}'.format(self._prior)
Martin Reinecke's avatar
Martin Reinecke committed
461
        return 'StandardHamiltonian:\n' + utilities.indent(subs)
Philipp Arras's avatar
Philipp Arras committed
462

463
464
465
466
    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
467

Martin Reinecke's avatar
Martin Reinecke committed
468
class AveragedEnergy(EnergyOperator):
Philipp Arras's avatar
Docs    
Philipp Arras committed
469
    """Averages an energy over samples.
Martin Reinecke's avatar
Martin Reinecke committed
470

471
472
473
    Parameters
    ----------
    h: Hamiltonian
Philipp Arras's avatar
Philipp Arras committed
474
       The energy to be averaged.
Martin Reinecke's avatar
Martin Reinecke committed
475
    res_samples : iterable of Fields
Torsten Ensslin's avatar
Torsten Ensslin committed
476
477
       Set of residual sample points to be added to mean field for
       approximate estimation of the KL.
478

Philipp Arras's avatar
Docs    
Philipp Arras committed
479
480
481
482
483
    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
484

Philipp Arras's avatar
Docs    
Philipp Arras committed
485
486
487
    - :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
488
    """
Martin Reinecke's avatar
Martin Reinecke committed
489
490
491

    def __init__(self, h, res_samples):
        self._h = h
Martin Reinecke's avatar
Martin Reinecke committed
492
        self._domain = h.domain
Martin Reinecke's avatar
Martin Reinecke committed
493
494
        self._res_samples = tuple(res_samples)

Philipp Arras's avatar
Philipp Arras committed
495
    def apply(self, x):
496
        self._check_input(x)
Philipp Arras's avatar
Philipp Arras committed
497
498
        mymap = map(lambda v: self._h(x+v), self._res_samples)
        return utilities.my_sum(mymap)/len(self._res_samples)