field.py 47.5 KB
Newer Older
1
2
3
4
5
6
7
8
9
10
11
12
# 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/>.
Theo Steininger's avatar
Theo Steininger committed
13
14
15
16
17
#
# Copyright(C) 2013-2017 Max-Planck-Society
#
# NIFTy is being developed at the Max-Planck-Institut fuer Astrophysik
# and financially supported by the Studienstiftung des deutschen Volkes.
18

csongor's avatar
csongor committed
19
from __future__ import division
Martin Reinecke's avatar
Martin Reinecke committed
20
21
22
from builtins import zip
from builtins import str
from builtins import range
23
24

import itertools
csongor's avatar
csongor committed
25
26
import numpy as np

Theo Steininger's avatar
Theo Steininger committed
27
28
from keepers import Versionable,\
                    Loggable
Jait Dixit's avatar
Jait Dixit committed
29

30
from d2o import distributed_data_object,\
31
    STRATEGIES as DISTRIBUTION_STRATEGIES
csongor's avatar
csongor committed
32

Martin Reinecke's avatar
Martin Reinecke committed
33
from .config import nifty_configuration as gc
csongor's avatar
csongor committed
34

Martin Reinecke's avatar
Martin Reinecke committed
35
from .domain_object import DomainObject
36

Martin Reinecke's avatar
Martin Reinecke committed
37
from .spaces.power_space import PowerSpace
csongor's avatar
csongor committed
38

Martin Reinecke's avatar
Martin Reinecke committed
39
40
from . import nifty_utilities as utilities
from .random import Random
Martin Reinecke's avatar
Martin Reinecke committed
41
from functools import reduce
42

csongor's avatar
csongor committed
43

Jait Dixit's avatar
Jait Dixit committed
44
class Field(Loggable, Versionable, object):
Theo Steininger's avatar
Theo Steininger committed
45
46
47
    """ The discrete representation of a continuous field over multiple spaces.

    In NIFTY, Fields are used to store data arrays and carry all the needed
48
    metainformation (i.e. the domain) for operators to be able to work on them.
Theo Steininger's avatar
Theo Steininger committed
49
50
    In addition Field has methods to work with power-spectra.

51
52
53
54
    Parameters
    ----------
    domain : DomainObject
        One of the space types NIFTY supports. RGSpace, GLSpace, HPSpace,
Theo Steininger's avatar
Theo Steininger committed
55
        LMSpace or PowerSpace. It might also be a FieldArray, which is
56
        an unstructured domain.
Theo Steininger's avatar
Theo Steininger committed
57

58
59
60
61
    val : scalar, numpy.ndarray, distributed_data_object, Field
        The values the array should contain after init. A scalar input will
        fill the whole array with this scalar. If an array is provided the
        array's dimensions must match the domain's.
Theo Steininger's avatar
Theo Steininger committed
62

63
64
    dtype : type
        A numpy.type. Most common are int, float and complex.
Theo Steininger's avatar
Theo Steininger committed
65

66
67
68
69
70
71
    distribution_strategy: optional[{'fftw', 'equal', 'not', 'freeform'}]
        Specifies which distributor will be created and used.
        'fftw'      uses the distribution strategy of pyfftw,
        'equal'     tries to  distribute the data as uniform as possible
        'not'       does not distribute the data at all
        'freeform'  distribute the data according to the given local data/shape
Theo Steininger's avatar
Theo Steininger committed
72

73
74
75
76
77
    copy: boolean

    Attributes
    ----------
    val : distributed_data_object
Theo Steininger's avatar
Theo Steininger committed
78

79
80
81
82
83
84
85
    domain : DomainObject
        See Parameters.
    domain_axes : tuple of tuples
        Enumerates the axes of the Field
    dtype : type
        Contains the datatype stored in the Field.
    distribution_strategy : string
Theo Steininger's avatar
Theo Steininger committed
86
87
        Name of the used distribution_strategy.

88
89
90
91
92
93
94
    Raise
    -----
    TypeError
        Raised if
            *the given domain contains something that is not a DomainObject
             instance
            *val is an array that has a different dimension than the domain
Theo Steininger's avatar
Theo Steininger committed
95

96
97
98
99
100
101
102
103
104
105
106
    Examples
    --------
    >>> a = Field(RGSpace([4,5]),val=2)
    >>> a.val
    <distributed_data_object>
    array([[2, 2, 2, 2, 2],
           [2, 2, 2, 2, 2],
           [2, 2, 2, 2, 2],
           [2, 2, 2, 2, 2]])
    >>> a.dtype
    dtype('int64')
Theo Steininger's avatar
Theo Steininger committed
107

108
109
110
111
112
    See Also
    --------
    distributed_data_object

    """
113

Theo Steininger's avatar
Theo Steininger committed
114
    # ---Initialization methods---
115

116
    def __init__(self, domain=None, val=None, dtype=None,
117
                 distribution_strategy=None, copy=False):
csongor's avatar
csongor committed
118

119
        self.domain = self._parse_domain(domain=domain, val=val)
120
        self.domain_axes = self._get_axes_tuple(self.domain)
csongor's avatar
csongor committed
121

Theo Steininger's avatar
Theo Steininger committed
122
        self.dtype = self._infer_dtype(dtype=dtype,
123
                                       val=val)
124

125
126
127
        self.distribution_strategy = self._parse_distribution_strategy(
                                distribution_strategy=distribution_strategy,
                                val=val)
csongor's avatar
csongor committed
128

129
130
131
132
        if val is None:
            self._val = None
        else:
            self.set_val(new_val=val, copy=copy)
csongor's avatar
csongor committed
133

134
    def _parse_domain(self, domain, val=None):
135
        if domain is None:
136
137
138
139
            if isinstance(val, Field):
                domain = val.domain
            else:
                domain = ()
140
        elif isinstance(domain, DomainObject):
141
            domain = (domain,)
142
143
144
        elif not isinstance(domain, tuple):
            domain = tuple(domain)

csongor's avatar
csongor committed
145
        for d in domain:
146
            if not isinstance(d, DomainObject):
147
148
                raise TypeError(
                    "Given domain contains something that is not a "
149
                    "DomainObject instance.")
csongor's avatar
csongor committed
150
151
        return domain

Theo Steininger's avatar
Theo Steininger committed
152
153
154
155
156
157
158
159
160
161
    def _get_axes_tuple(self, things_with_shape, start=0):
        i = start
        axes_list = []
        for thing in things_with_shape:
            l = []
            for j in range(len(thing.shape)):
                l += [i]
                i += 1
            axes_list += [tuple(l)]
        return tuple(axes_list)
162

163
    def _infer_dtype(self, dtype, val):
csongor's avatar
csongor committed
164
        if dtype is None:
165
            try:
166
                dtype = val.dtype
167
            except AttributeError:
Theo Steininger's avatar
Theo Steininger committed
168
169
170
                try:
                    if val is None:
                        raise TypeError
171
                    dtype = np.result_type(val)
Theo Steininger's avatar
Theo Steininger committed
172
                except(TypeError):
173
                    dtype = np.dtype(gc['default_field_dtype'])
Theo Steininger's avatar
Theo Steininger committed
174
        else:
175
            dtype = np.dtype(dtype)
176

177
178
        dtype = np.result_type(dtype, np.float)

Theo Steininger's avatar
Theo Steininger committed
179
        return dtype
180

181
182
    def _parse_distribution_strategy(self, distribution_strategy, val):
        if distribution_strategy is None:
183
            if isinstance(val, distributed_data_object):
184
                distribution_strategy = val.distribution_strategy
185
            elif isinstance(val, Field):
186
                distribution_strategy = val.distribution_strategy
187
            else:
188
                self.logger.debug("distribution_strategy set to default!")
189
                distribution_strategy = gc['default_distribution_strategy']
190
        elif distribution_strategy not in DISTRIBUTION_STRATEGIES['global']:
191
192
193
            raise ValueError(
                    "distribution_strategy must be a global-type "
                    "strategy.")
194
        return distribution_strategy
195
196

    # ---Factory methods---
197

198
    @classmethod
199
    def from_random(cls, random_type, domain=None, dtype=None,
200
                    distribution_strategy=None, **kwargs):
201
202
203
204
205
        """ Draws a random field with the given parameters.

        Parameters
        ----------
        cls : class
Theo Steininger's avatar
Theo Steininger committed
206

207
208
209
        random_type : String
            'pm1', 'normal', 'uniform' are the supported arguments for this
            method.
Theo Steininger's avatar
Theo Steininger committed
210

211
212
        domain : DomainObject
            The domain of the output random field
Theo Steininger's avatar
Theo Steininger committed
213

214
215
        dtype : type
            The datatype of the output random field
Theo Steininger's avatar
Theo Steininger committed
216

217
218
        distribution_strategy : all supported distribution strategies
            The distribution strategy of the output random field
Theo Steininger's avatar
Theo Steininger committed
219

220
221
222
223
224
225
226
        Returns
        -------
        out : Field
            The output object.

        See Also
        --------
227
        power_synthesize
Theo Steininger's avatar
Theo Steininger committed
228

229
230

        """
Theo Steininger's avatar
Theo Steininger committed
231

232
        # create a initially empty field
233
        f = cls(domain=domain, dtype=dtype,
234
                distribution_strategy=distribution_strategy)
235
236
237
238
239
240
241

        # now use the processed input in terms of f in order to parse the
        # random arguments
        random_arguments = cls._parse_random_arguments(random_type=random_type,
                                                       f=f,
                                                       **kwargs)

Martin Reinecke's avatar
Martin Reinecke committed
242
        # extract the distributed_data_object from f and apply the appropriate
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
        # random number generator to it
        sample = f.get_val(copy=False)
        generator_function = getattr(Random, random_type)
        sample.apply_generator(
            lambda shape: generator_function(dtype=f.dtype,
                                             shape=shape,
                                             **random_arguments))
        return f

    @staticmethod
    def _parse_random_arguments(random_type, f, **kwargs):
        if random_type == "pm1":
            random_arguments = {}

        elif random_type == "normal":
            mean = kwargs.get('mean', 0)
            std = kwargs.get('std', 1)
            random_arguments = {'mean': mean,
                                'std': std}

        elif random_type == "uniform":
            low = kwargs.get('low', 0)
            high = kwargs.get('high', 1)
            random_arguments = {'low': low,
                                'high': high}

csongor's avatar
csongor committed
269
        else:
270
271
            raise KeyError(
                "unsupported random key '" + str(random_type) + "'.")
csongor's avatar
csongor committed
272

273
        return random_arguments
csongor's avatar
csongor committed
274

275
276
    # ---Powerspectral methods---

Martin Reinecke's avatar
Martin Reinecke committed
277
    def power_analyze(self, spaces=None, logarithmic=None, nbin=None,
278
                      binbounds=None, keep_phase_information=False):
Theo Steininger's avatar
Theo Steininger committed
279
        """ Computes the square root power spectrum for a subspace of `self`.
Theo Steininger's avatar
Theo Steininger committed
280

Theo Steininger's avatar
Theo Steininger committed
281
282
283
        Creates a PowerSpace for the space addressed by `spaces` with the given
        binning and computes the power spectrum as a Field over this
        PowerSpace. This can only be done if the subspace to  be analyzed is a
284
        harmonic space. The resulting field has the same units as the initial
Theo Steininger's avatar
Theo Steininger committed
285
        field, corresponding to the square root of the power spectrum.
286
287
288

        Parameters
        ----------
Theo Steininger's avatar
Theo Steininger committed
289
290
291
292
293
        spaces : int *optional*
            The subspace for which the powerspectrum shall be computed
            (default : None).
        logarithmic : boolean *optional*
            True if the output PowerSpace should use logarithmic binning.
Martin Reinecke's avatar
Martin Reinecke committed
294
            {default : None}
Theo Steininger's avatar
Theo Steininger committed
295
296
297
298
299
300
        nbin : int *optional*
            The number of bins the resulting PowerSpace shall have
            (default : None).
            if nbin==None : maximum number of bins is used
        binbounds : array-like *optional*
            Inner bounds of the bins (default : None).
Martin Reinecke's avatar
Martin Reinecke committed
301
302
            Overrides nbin and logarithmic.
            if binbounds==None : bins are inferred.
303
304
305
306
307
308
309
310
311
312
        keep_phase_information : boolean, *optional*
            If False, return a real-valued result containing the power spectrum
            of the input Field.
            If True, return a complex-valued result whose real component
            contains the power spectrum computed from the real part of the
            input Field, and whose imaginary component contains the power
            spectrum computed from the imaginary part of the input Field.
            The absolute value of this result should be identical to the output
            of power_analyze with keep_phase_information=False.
            (default : False).
Theo Steininger's avatar
Theo Steininger committed
313

314
315
316
317
        Raise
        -----
        ValueError
            Raised if
Theo Steininger's avatar
Theo Steininger committed
318
319
                *len(domain) is != 1 when spaces==None
                *len(spaces) is != 1 if not None
320
                *the analyzed space is not harmonic
Theo Steininger's avatar
Theo Steininger committed
321

322
323
        Returns
        -------
Theo Steininger's avatar
Theo Steininger committed
324
        out : Field
325
326
327
328
329
330
            The output object. It's domain is a PowerSpace and it contains
            the power spectrum of 'self's field.

        See Also
        --------
        power_synthesize, PowerSpace
Theo Steininger's avatar
Theo Steininger committed
331

332
        """
Theo Steininger's avatar
Theo Steininger committed
333

Theo Steininger's avatar
Theo Steininger committed
334
        # check if all spaces in `self.domain` are either harmonic or
335
336
337
        # power_space instances
        for sp in self.domain:
            if not sp.harmonic and not isinstance(sp, PowerSpace):
Theo Steininger's avatar
Theo Steininger committed
338
                self.logger.info(
339
                    "Field has a space in `domain` which is neither "
340
341
342
                    "harmonic nor a PowerSpace.")

        # check if the `spaces` input is valid
343
344
        spaces = utilities.cast_axis_to_tuple(spaces, len(self.domain))
        if spaces is None:
Martin Reinecke's avatar
Martin Reinecke committed
345
            spaces = list(range(len(self.domain)))
346
347

        if len(spaces) == 0:
348
349
            raise ValueError(
                "No space for analysis specified.")
350

351
352
353
354
355
356
357
358
359
360
361
362
363
        if keep_phase_information:
            parts_val = self._hermitian_decomposition(
                                              domain=self.domain,
                                              val=self.val,
                                              spaces=spaces,
                                              domain_axes=self.domain_axes,
                                              preserve_gaussian_variance=False)
            parts = [self.copy_empty().set_val(part_val, copy=False)
                     for part_val in parts_val]
        else:
            parts = [self]

        parts = [abs(part)**2 for part in parts]
364
365

        for space_index in spaces:
366
367
            parts = [self._single_power_analyze(
                                work_field=part,
368
369
370
                                space_index=space_index,
                                logarithmic=logarithmic,
                                nbin=nbin,
371
372
                                binbounds=binbounds)
                     for part in parts]
373

374
375
376
377
378
379
        if keep_phase_information:
            result_field = parts[0] + 1j*parts[1]
        else:
            result_field = parts[0]

        return result_field
380
381
382

    @classmethod
    def _single_power_analyze(cls, work_field, space_index, logarithmic, nbin,
383
                              binbounds):
384

385
        if not work_field.domain[space_index].harmonic:
386
387
            raise ValueError(
                "The analyzed space must be harmonic.")
388

389
390
391
392
393
394
        # Create the target PowerSpace instance:
        # If the associated signal-space field was real, we extract the
        # hermitian and anti-hermitian parts of `self` and put them
        # into the real and imaginary parts of the power spectrum.
        # If it was complex, all the power is put into a real power spectrum.

395
        distribution_strategy = \
396
397
            work_field.val.get_axes_local_distribution_strategy(
                work_field.domain_axes[space_index])
398

399
        harmonic_domain = work_field.domain[space_index]
400
        power_domain = PowerSpace(harmonic_partner=harmonic_domain,
401
                                  distribution_strategy=distribution_strategy,
Theo Steininger's avatar
Theo Steininger committed
402
403
                                  logarithmic=logarithmic, nbin=nbin,
                                  binbounds=binbounds)
404

405
406
        power_spectrum = cls._calculate_power_spectrum(
                                field_val=work_field.val,
Martin Reinecke's avatar
Martin Reinecke committed
407
                                pdomain=power_domain,
408
                                axes=work_field.domain_axes[space_index])
409
410

        # create the result field and put power_spectrum into it
411
        result_domain = list(work_field.domain)
412
        result_domain[space_index] = power_domain
413
        result_dtype = power_spectrum.dtype
414

415
        result_field = work_field.copy_empty(
416
                   domain=result_domain,
417
                   dtype=result_dtype,
418
                   distribution_strategy=power_spectrum.distribution_strategy)
419
420
421
422
        result_field.set_val(new_val=power_spectrum, copy=False)

        return result_field

423
    @classmethod
Martin Reinecke's avatar
Martin Reinecke committed
424
    def _calculate_power_spectrum(cls, field_val, pdomain, axes=None):
425

Martin Reinecke's avatar
Martin Reinecke committed
426
427
428
        pindex = pdomain.pindex
        # MR FIXME: how about iterating over slices, instead of replicating
        # pindex? Would save memory and probably isn't slower.
429
        if axes is not None:
430
431
432
433
434
435
            pindex = cls._shape_up_pindex(
                            pindex=pindex,
                            target_shape=field_val.shape,
                            target_strategy=field_val.distribution_strategy,
                            axes=axes)
        power_spectrum = pindex.bincount(weights=field_val,
436
                                         axis=axes)
Martin Reinecke's avatar
Martin Reinecke committed
437
        rho = pdomain.rho
438
439
440
441
442
443
444
445
        if axes is not None:
            new_rho_shape = [1, ] * len(power_spectrum.shape)
            new_rho_shape[axes[0]] = len(rho)
            rho = rho.reshape(new_rho_shape)
        power_spectrum /= rho

        return power_spectrum

446
447
    @staticmethod
    def _shape_up_pindex(pindex, target_shape, target_strategy, axes):
448
449
        if pindex.distribution_strategy not in \
                DISTRIBUTION_STRATEGIES['global']:
450
            raise ValueError("pindex's distribution strategy must be "
451
452
453
454
455
456
                             "global-type")

        if pindex.distribution_strategy in DISTRIBUTION_STRATEGIES['slicing']:
            if ((0 not in axes) or
                    (target_strategy is not pindex.distribution_strategy)):
                raise ValueError(
457
                    "A slicing distributor shall not be reshaped to "
458
459
460
461
462
463
464
465
466
467
468
469
470
                    "something non-sliced.")

        semiscaled_shape = [1, ] * len(target_shape)
        for i in axes:
            semiscaled_shape[i] = target_shape[i]
        local_data = pindex.get_local_data(copy=False)
        semiscaled_local_data = local_data.reshape(semiscaled_shape)
        result_obj = pindex.copy_empty(global_shape=target_shape,
                                       distribution_strategy=target_strategy)
        result_obj.set_full_data(semiscaled_local_data, copy=False)

        return result_obj

471
472
    def power_synthesize(self, spaces=None, real_power=True, real_signal=True,
                         mean=None, std=None):
Theo Steininger's avatar
Theo Steininger committed
473
        """ Yields a sampled field with `self`**2 as its power spectrum.
Theo Steininger's avatar
Theo Steininger committed
474

Theo Steininger's avatar
Theo Steininger committed
475
476
        This method draws a Gaussian random field in the harmonic partner
        domain of this fields domains, using this field as power spectrum.
Theo Steininger's avatar
Theo Steininger committed
477

478
479
480
        Parameters
        ----------
        spaces : {tuple, int, None} *optional*
Theo Steininger's avatar
Theo Steininger committed
481
482
483
            Specifies the subspace containing all the PowerSpaces which
            should be converted (default : None).
            if spaces==None : Tries to convert the whole domain.
484
        real_power : boolean *optional*
Theo Steininger's avatar
Theo Steininger committed
485
486
            Determines whether the power spectrum is treated as intrinsically
            real or complex (default : True).
487
        real_signal : boolean *optional*
Theo Steininger's avatar
Theo Steininger committed
488
489
490
491
492
493
            True will result in a purely real signal-space field
            (default : True).
        mean : float *optional*
            The mean of the Gaussian noise field which is used for the Field
            synthetization (default : None).
            if mean==None : mean will be set to 0
494
        std : float *optional*
Theo Steininger's avatar
Theo Steininger committed
495
496
497
            The standard deviation of the Gaussian noise field which is used
            for the Field synthetization (default : None).
            if std==None : std will be set to 1
Theo Steininger's avatar
Theo Steininger committed
498

499
500
501
502
        Returns
        -------
        out : Field
            The output object. A random field created with the power spectrum
Theo Steininger's avatar
Theo Steininger committed
503
            stored in the `spaces` in `self`.
504

Theo Steininger's avatar
Theo Steininger committed
505
506
507
508
509
510
        Notes
        -----
        For this the spaces specified by `spaces` must be a PowerSpace.
        This expects this field to be the square root of a power spectrum, i.e.
        to have the unit of the field to be sampled.

511
512
513
        See Also
        --------
        power_analyze
Theo Steininger's avatar
Theo Steininger committed
514
515
516
517
518

        Raises
        ------
        ValueError : If domain specified by `spaces` is not a PowerSpace.

519
        """
Theo Steininger's avatar
Theo Steininger committed
520

521
522
523
        # check if the `spaces` input is valid
        spaces = utilities.cast_axis_to_tuple(spaces, len(self.domain))

Theo Steininger's avatar
Theo Steininger committed
524
        if spaces is None:
Martin Reinecke's avatar
Martin Reinecke committed
525
            spaces = list(range(len(self.domain)))
Theo Steininger's avatar
Theo Steininger committed
526

527
528
529
530
531
        for power_space_index in spaces:
            power_space = self.domain[power_space_index]
            if not isinstance(power_space, PowerSpace):
                raise ValueError("A PowerSpace is needed for field "
                                 "synthetization.")
532
533
534

        # create the result domain
        result_domain = list(self.domain)
535
536
        for power_space_index in spaces:
            power_space = self.domain[power_space_index]
537
            harmonic_domain = power_space.harmonic_partner
538
            result_domain[power_space_index] = harmonic_domain
539
540
541

        # create random samples: one or two, depending on whether the
        # power spectrum is real or complex
542
        if real_power:
543
            result_list = [None]
544
545
        else:
            result_list = [None, None]
546

547
548
        result_list = [self.__class__.from_random(
                             'normal',
549
550
551
                             mean=mean,
                             std=std,
                             domain=result_domain,
552
                             dtype=np.complex,
553
                             distribution_strategy=self.distribution_strategy)
554
555
556
557
558
559
                       for x in result_list]

        # from now on extract the values from the random fields for further
        # processing without killing the fields.
        # if the signal-space field should be real, hermitianize the field
        # components
560
561

        spec = self.val.get_full_data()
562
563
        spec = np.sqrt(spec)

564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
        for power_space_index in spaces:
            spec = self._spec_to_rescaler(spec, result_list, power_space_index)
        local_rescaler = spec

        result_val_list = [x.val for x in result_list]

        # apply the rescaler to the random fields
        result_val_list[0].apply_scalar_function(
                                            lambda x: x * local_rescaler.real,
                                            inplace=True)

        if not real_power:
            result_val_list[1].apply_scalar_function(
                                            lambda x: x * local_rescaler.imag,
                                            inplace=True)

580
        if real_signal:
581
            result_val_list = [self._hermitian_decomposition(
582
583
584
585
586
                                            result_domain,
                                            result_val,
                                            spaces,
                                            result_list[0].domain_axes,
                                            preserve_gaussian_variance=True)[0]
587
                               for result_val in result_val_list]
588
589
590
591
592
593
594

        # store the result into the fields
        [x.set_val(new_val=y, copy=False) for x, y in
            zip(result_list, result_val_list)]

        if real_power:
            result = result_list[0]
595
        else:
596
597
598
599
            result = result_list[0] + 1j*result_list[1]

        return result

600
    @staticmethod
601
602
    def _hermitian_decomposition(domain, val, spaces, domain_axes,
                                 preserve_gaussian_variance=False):
603
604
605
606
607
608
609
610
611

        flipped_val = val
        for space in spaces:
            flipped_val = domain[space].hermitianize_inverter(
                                                    x=flipped_val,
                                                    axes=domain_axes[space])
        flipped_val = flipped_val.conjugate()
        h = (val + flipped_val)/2.
        a = val - h
612
613

        # correct variance
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
        if preserve_gaussian_variance:
            h *= np.sqrt(2)
            a *= np.sqrt(2)

            if not issubclass(val.dtype.type, np.complexfloating):
                # in principle one must not correct the variance for the fixed
                # points of the hermitianization. However, for a complex field
                # the input field loses half of its power at its fixed points
                # in the `hermitian` part. Hence, here a factor of sqrt(2) is
                # also necessary!
                # => The hermitianization can be done on a space level since
                # either nothing must be done (LMSpace) or ALL points need a
                # factor of sqrt(2)
                # => use the preserve_gaussian_variance flag in the
                # hermitian_decomposition method above.

                # This code is for educational purposes:
                fixed_points = [domain[i].hermitian_fixed_points()
                                for i in spaces]
                fixed_points = [[fp] if fp is None else fp
                                for fp in fixed_points]

                for product_point in itertools.product(*fixed_points):
                    slice_object = np.array((slice(None), )*len(val.shape),
                                            dtype=np.object)
                    for i, sp in enumerate(spaces):
                        point_component = product_point[i]
                        if point_component is None:
                            point_component = slice(None)
                        slice_object[list(domain_axes[sp])] = point_component

                    slice_object = tuple(slice_object)
                    h[slice_object] /= np.sqrt(2)
                    a[slice_object] /= np.sqrt(2)
648
649
        return (h, a)

650
651
    def _spec_to_rescaler(self, spec, result_list, power_space_index):
        power_space = self.domain[power_space_index]
652
653
654

        # weight the random fields with the power spectrum
        # therefore get the pindex from the power space
655
        pindex = power_space.pindex
656
657
658
659
660
661
662
        # take the local data from pindex. This data must be compatible to the
        # local data of the field given the slice of the PowerSpace
        local_distribution_strategy = \
            result_list[0].val.get_axes_local_distribution_strategy(
                result_list[0].domain_axes[power_space_index])

        if pindex.distribution_strategy is not local_distribution_strategy:
663
            self.logger.warn(
Martin Reinecke's avatar
Martin Reinecke committed
664
                "The distribution_strategy of pindex does not fit the "
665
666
667
668
669
670
671
                "slice_local distribution strategy of the synthesized field.")

        # Now use numpy advanced indexing in order to put the entries of the
        # power spectrum into the appropriate places of the pindex array.
        # Do this for every 'pindex-slice' in parallel using the 'slice(None)'s
        local_pindex = pindex.get_local_data(copy=False)

672
673
674
675
676
        local_blow_up = [slice(None)]*len(spec.shape)
        # it is important to count from behind, since spec potentially grows
        # with every iteration
        index = self.domain_axes[power_space_index][0]-len(self.shape)
        local_blow_up[index] = local_pindex
677
        # here, the power_spectrum is distributed into the new shape
678
679
        local_rescaler = spec[local_blow_up]
        return local_rescaler
680

Theo Steininger's avatar
Theo Steininger committed
681
    # ---Properties---
682

Theo Steininger's avatar
Theo Steininger committed
683
    def set_val(self, new_val=None, copy=False):
Theo Steininger's avatar
Theo Steininger committed
684
        """ Sets the fields distributed_data_object.
685
686
687

        Parameters
        ----------
Theo Steininger's avatar
Theo Steininger committed
688
        new_val : scalar, array-like, Field, None *optional*
689
690
            The values to be stored in the field.
            {default : None}
Theo Steininger's avatar
Theo Steininger committed
691

692
        copy : boolean, *optional*
Theo Steininger's avatar
Theo Steininger committed
693
694
            If False, Field tries to not copy the input data but use it
            directly.
695
696
697
698
699
700
            {default : False}
        See Also
        --------
        val

        """
Theo Steininger's avatar
Theo Steininger committed
701

702
703
        new_val = self.cast(new_val)
        if copy:
Theo Steininger's avatar
Theo Steininger committed
704
705
            new_val = new_val.copy()
        self._val = new_val
706
        return self
csongor's avatar
csongor committed
707

708
    def get_val(self, copy=False):
Theo Steininger's avatar
Theo Steininger committed
709
        """ Returns the distributed_data_object associated with this Field.
710
711
712
713

        Parameters
        ----------
        copy : boolean
Theo Steininger's avatar
Theo Steininger committed
714
715
            If true, a copy of the Field's underlying distributed_data_object
            is returned.
Theo Steininger's avatar
Theo Steininger committed
716

717
718
719
720
721
722
723
724
725
        Returns
        -------
        out : distributed_data_object

        See Also
        --------
        val

        """
Theo Steininger's avatar
Theo Steininger committed
726

727
728
729
        if self._val is None:
            self.set_val(None)

730
        if copy:
Theo Steininger's avatar
Theo Steininger committed
731
            return self._val.copy()
732
        else:
Theo Steininger's avatar
Theo Steininger committed
733
            return self._val
csongor's avatar
csongor committed
734

Theo Steininger's avatar
Theo Steininger committed
735
736
    @property
    def val(self):
Theo Steininger's avatar
Theo Steininger committed
737
        """ Returns the distributed_data_object associated with this Field.
Theo Steininger's avatar
Theo Steininger committed
738

739
740
741
742
743
744
745
746
747
        Returns
        -------
        out : distributed_data_object

        See Also
        --------
        get_val

        """
Theo Steininger's avatar
Theo Steininger committed
748

749
        return self.get_val(copy=False)
csongor's avatar
csongor committed
750

Theo Steininger's avatar
Theo Steininger committed
751
752
    @val.setter
    def val(self, new_val):
753
        self.set_val(new_val=new_val, copy=False)
csongor's avatar
csongor committed
754

755
756
    @property
    def shape(self):
Theo Steininger's avatar
Theo Steininger committed
757
        """ Returns the total shape of the Field's data array.
Theo Steininger's avatar
Theo Steininger committed
758

759
760
761
        Returns
        -------
        out : tuple
Martin Reinecke's avatar
Martin Reinecke committed
762
            The output object. The tuple contains the dimensions of the spaces
763
764
765
766
767
768
769
            in domain.

        See Also
        --------
        dim

        """
Theo Steininger's avatar
Theo Steininger committed
770

771
        shape_tuple = tuple(sp.shape for sp in self.domain)
772
773
774
775
        try:
            global_shape = reduce(lambda x, y: x + y, shape_tuple)
        except TypeError:
            global_shape = ()
csongor's avatar
csongor committed
776

777
        return global_shape
csongor's avatar
csongor committed
778

779
780
    @property
    def dim(self):
Theo Steininger's avatar
Theo Steininger committed
781
        """ Returns the total number of pixel-dimensions the field has.
Theo Steininger's avatar
Theo Steininger committed
782

Theo Steininger's avatar
Theo Steininger committed
783
        Effectively, all values from shape are multiplied.
Theo Steininger's avatar
Theo Steininger committed
784

785
786
787
788
789
790
791
792
793
794
        Returns
        -------
        out : int
            The dimension of the Field.

        See Also
        --------
        shape

        """
Theo Steininger's avatar
Theo Steininger committed
795

796
        dim_tuple = tuple(sp.dim for sp in self.domain)
Theo Steininger's avatar
Theo Steininger committed
797
        try:
Martin Reinecke's avatar
Martin Reinecke committed
798
            return int(reduce(lambda x, y: x * y, dim_tuple))
Theo Steininger's avatar
Theo Steininger committed
799
800
        except TypeError:
            return 0
csongor's avatar
csongor committed
801

802
803
    @property
    def dof(self):
Theo Steininger's avatar
Theo Steininger committed
804
805
806
807
808
809
        """ Returns the total number of degrees of freedom the Field has. For
        real Fields this is equal to `self.dim`. For complex Fields it is
        2*`self.dim`.

        """

Theo Steininger's avatar
Theo Steininger committed
810
811
812
813
814
815
816
        dof = self.dim
        if issubclass(self.dtype.type, np.complexfloating):
            dof *= 2
        return dof

    @property
    def total_volume(self):
Theo Steininger's avatar
Theo Steininger committed
817
818
819
        """ Returns the total volume of all spaces in the domain.
        """

Theo Steininger's avatar
Theo Steininger committed
820
        volume_tuple = tuple(sp.total_volume for sp in self.domain)
821
        try:
Theo Steininger's avatar
Theo Steininger committed
822
            return reduce(lambda x, y: x * y, volume_tuple)
823
        except TypeError:
Theo Steininger's avatar
Theo Steininger committed
824
            return 0.
825

Theo Steininger's avatar
Theo Steininger committed
826
    # ---Special unary/binary operations---
827

csongor's avatar
csongor committed
828
    def cast(self, x=None, dtype=None):
Theo Steininger's avatar
Theo Steininger committed
829
        """ Transforms x to a d2o with the correct dtype and shape.
Theo Steininger's avatar
Theo Steininger committed
830

831
832
        Parameters
        ----------
Theo Steininger's avatar
Theo Steininger committed
833
        x : scalar, d2o, Field, array_like
834
835
            The input that shall be casted on a d2o of the same shape like the
            domain.
Theo Steininger's avatar
Theo Steininger committed
836

837
        dtype : type
Theo Steininger's avatar
Theo Steininger committed
838
839
            The datatype the output shall have. This can be used to override
            the fields dtype.
Theo Steininger's avatar
Theo Steininger committed
840

841
842
843
844
845
846
847
848
849
850
        Returns
        -------
        out : distributed_data_object
            The output object.

        See Also
        --------
        _actual_cast

        """
csongor's avatar
csongor committed
851
852
        if dtype is None:
            dtype = self.dtype
853
854
        else:
            dtype = np.dtype(dtype)
855

856
857
        casted_x = x

858
        for ind, sp in enumerate(self.domain):
859
            casted_x = sp.pre_cast(casted_x,
860
861
862
                                   axes=self.domain_axes[ind])

        casted_x = self._actual_cast(casted_x, dtype=dtype)
863
864

        for ind, sp in enumerate(self.domain):
865
866
            casted_x = sp.post_cast(casted_x,
                                    axes=self.domain_axes[ind])
867

868
        return casted_x
csongor's avatar
csongor committed
869

Theo Steininger's avatar
Theo Steininger committed
870
    def _actual_cast(self, x, dtype=None):
871
        if isinstance(x, Field):
csongor's avatar
csongor committed
872
873
874
875
876
            x = x.get_val()

        if dtype is None:
            dtype = self.dtype

877
        return_x = distributed_data_object(
878
879
880
                            global_shape=self.shape,
                            dtype=dtype,
                            distribution_strategy=self.distribution_strategy)
881
882
        return_x.set_full_data(x, copy=False)
        return return_x
Theo Steininger's avatar
Theo Steininger committed
883

884
    def copy(self, domain=None, dtype=None, distribution_strategy=None):
885
        """ Returns a full copy of the Field.
Theo Steininger's avatar
Theo Steininger committed
886

887
888
889
890
891
892
893
894
895
        If no keyword arguments are given, the returned object will be an
        identical copy of the original Field. By explicit specification one is
        able to define the domain, the dtype and the distribution_strategy of
        the returned Field.

        Parameters
        ----------
        domain : DomainObject
            The new domain the Field shall have.
Theo Steininger's avatar
Theo Steininger committed
896

897
898
        dtype : type
            The new dtype the Field shall have.
Theo Steininger's avatar
Theo Steininger committed
899

900
        distribution_strategy : all supported distribution strategies
Theo Steininger's avatar
Theo Steininger committed
901
902
            The new distribution strategy the Field shall have.

903
904
905
906
907
908
909
910
911
912
        Returns
        -------
        out : Field
            The output object. An identical copy of 'self'.

        See Also
        --------
        copy_empty

        """
Theo Steininger's avatar
Theo Steininger committed
913

Theo Steininger's avatar
Theo Steininger committed
914
        copied_val = self.get_val(copy=True)
915
916
917
918
        new_field = self.copy_empty(
                                domain=domain,
                                dtype=dtype,
                                distribution_strategy=distribution_strategy)
Theo Steininger's avatar
Theo Steininger committed
919
920
        new_field.set_val(new_val=copied_val, copy=False)
        return new_field
csongor's avatar
csongor committed
921

922
    def copy_empty(self, domain=None, dtype=None, distribution_strategy=None):
923
924
925
        """ Returns an empty copy of the Field.

        If no keyword arguments are given, the returned object will be an
Theo Steininger's avatar
Theo Steininger committed
926
927
928
929
930
        identical copy of the original Field. The memory for the data array
        is only allocated but not actively set to any value
        (c.f. numpy.ndarray.copy_empty). By explicit specification one is able
        to change the domain, the dtype and the distribution_strategy of the
        returned Field.
Theo Steininger's avatar
Theo Steininger committed
931

932
933
934
935
        Parameters
        ----------
        domain : DomainObject
            The new domain the Field shall have.
Theo Steininger's avatar
Theo Steininger committed
936

937
938
        dtype : type
            The new dtype the Field shall have.
Theo Steininger's avatar
Theo Steininger committed
939

Theo Steininger's avatar
Theo Steininger committed
940
        distribution_strategy : string, all supported distribution strategies
941
            The distribution strategy the new Field should have.
Theo Steininger's avatar
Theo Steininger committed
942

943
944
945
        Returns
        -------
        out : Field
Theo Steininger's avatar
Theo Steininger committed
946
            The output object.
947
948
949
950
951
952

        See Also
        --------
        copy

        """
Theo Steininger's avatar
Theo Steininger committed
953

Theo Steininger's avatar
Theo Steininger committed
954
955
        if domain is None:
            domain = self.domain
csongor's avatar
csongor committed
956
        else:
Theo Steininger's avatar
Theo Steininger committed
957
            domain = self._parse_domain(domain)
csongor's avatar
csongor committed
958

Theo Steininger's avatar
Theo Steininger committed
959
960
961
962
        if dtype is None:
            dtype = self.dtype
        else:
            dtype = np.dtype(dtype)
csongor's avatar
csongor committed
963

964
965
        if distribution_strategy is None:
            distribution_strategy = self.distribution_strategy
csongor's avatar
csongor committed
966

Theo Steininger's avatar
Theo Steininger committed
967
968
        fast_copyable = True
        try:
Martin Reinecke's avatar
Martin Reinecke committed
969
            for i in range(len(self.domain)):
Theo Steininger's avatar
Theo Steininger committed
970
971
972
973
974
975
976
                if self.domain[i] is not domain[i]:
                    fast_copyable = False
                    break
        except IndexError:
            fast_copyable = False

        if (fast_copyable and dtype == self.dtype and
977
                distribution_strategy == self.distribution_strategy):
Theo Steininger's avatar
Theo Steininger committed
978
979
980
981
            new_field = self._fast_copy_empty()
        else:
            new_field = Field(domain=domain,
                              dtype=dtype,
982
                              distribution_strategy=distribution_strategy)
Theo Steininger's avatar
Theo Steininger committed
983
        return new_field
csongor's avatar
csongor committed
984

Theo Steininger's avatar
Theo Steininger committed
985
986
987
988
989
990
    def _fast_copy_empty(self):
        # make an empty field
        new_field = EmptyField()
        # repair its class
        new_field.__class__ = self.__class__
        # copy domain, codomain and val
Martin Reinecke's avatar
Martin Reinecke committed
991
        for key, value in list(self.__dict__.items()):
992
            if key != '_val':
Theo Steininger's avatar
Theo Steininger committed
993
994
995
996
997
998
                new_field.__dict__[key] = value
            else:
                new_field.__dict__[key] = self.val.copy_empty()
        return new_field

    def weight(self, power=1, inplace=False, spaces=None):
Theo Steininger's avatar
Theo Steininger committed
999
        """ Weights the pixels of `self` with their invidual pixel-volume.
1000
1001
1002
1003

        Parameters
        ----------
        power : number
Theo Steininger's avatar
Theo Steininger committed
1004
            The pixels get weighted with the volume-factor**power.
Theo Steininger's avatar
Theo Steininger committed
1005

1006
        inplace : boolean
Theo Steininger's avatar
Theo Steininger committed
1007
1008
            If True, `self` will be weighted and returned. Otherwise, a copy
            is made.
Theo Steininger's avatar
Theo Steininger committed
1009

Theo Steininger's avatar
Theo Steininger committed
1010
1011
        spaces : tuple of ints
            Determines on which subspace the operation takes place.
Theo Steininger's avatar
Theo Steininger committed
1012

1013
1014
1015
        Returns
        -------
        out : Field
Theo Steininger's avatar
Theo Steininger committed
1016
            The weighted field.
1017
1018

        """