field.py 30.8 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
from builtins import zip
from builtins import range
22

23
import ast
csongor's avatar
csongor committed
24
25
import numpy as np

Martin Reinecke's avatar
Martin Reinecke committed
26
from .domain_object import DomainObject
27

Martin Reinecke's avatar
Martin Reinecke committed
28
from .spaces.power_space import PowerSpace
csongor's avatar
csongor committed
29

Martin Reinecke's avatar
Martin Reinecke committed
30
31
from . import nifty_utilities as utilities
from .random import Random
Martin Reinecke's avatar
Martin Reinecke committed
32
from functools import reduce
33

csongor's avatar
csongor committed
34

Martin Reinecke's avatar
Martin Reinecke committed
35
class Field(object):
Theo Steininger's avatar
Theo Steininger committed
36
37
38
    """ The discrete representation of a continuous field over multiple spaces.

    In NIFTY, Fields are used to store data arrays and carry all the needed
39
    metainformation (i.e. the domain) for operators to be able to work on them.
Theo Steininger's avatar
Theo Steininger committed
40
41
    In addition Field has methods to work with power-spectra.

42
43
44
45
    Parameters
    ----------
    domain : DomainObject
        One of the space types NIFTY supports. RGSpace, GLSpace, HPSpace,
Theo Steininger's avatar
Theo Steininger committed
46
        LMSpace or PowerSpace. It might also be a FieldArray, which is
47
        an unstructured domain.
Theo Steininger's avatar
Theo Steininger committed
48

Martin Reinecke's avatar
stage1    
Martin Reinecke committed
49
    val : scalar, numpy.ndarray, Field
50
51
52
        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
53

54
55
    dtype : type
        A numpy.type. Most common are int, float and complex.
Theo Steininger's avatar
Theo Steininger committed
56

57
58
59
60
    copy: boolean

    Attributes
    ----------
Martin Reinecke's avatar
stage1    
Martin Reinecke committed
61
    val : numpy.ndarray
Theo Steininger's avatar
Theo Steininger committed
62

63
64
65
66
67
68
    domain : DomainObject
        See Parameters.
    domain_axes : tuple of tuples
        Enumerates the axes of the Field
    dtype : type
        Contains the datatype stored in the Field.
Theo Steininger's avatar
Theo Steininger committed
69

70
71
72
73
74
75
76
    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
77

78
    """
79

Theo Steininger's avatar
Theo Steininger committed
80
    # ---Initialization methods---
81

Martin Reinecke's avatar
stage1    
Martin Reinecke committed
82
    def __init__(self, domain=None, val=None, dtype=None, copy=False):
83
        self.domain = self._parse_domain(domain=domain, val=val)
84
        self.domain_axes = self._get_axes_tuple(self.domain)
85

Martin Reinecke's avatar
Martin Reinecke committed
86
87
88
        shape_tuple = tuple(sp.shape for sp in self.domain)
        if len(shape_tuple)==0:
            global_shape = ()
89
        else:
Martin Reinecke's avatar
Martin Reinecke committed
90
91
            global_shape = reduce(lambda x, y: x + y, shape_tuple)
        dtype = self._infer_dtype(dtype=dtype, val=val)
Martin Reinecke's avatar
cleanup    
Martin Reinecke committed
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
        if isinstance(val, Field):
            if self.domain!=val.domain:
                raise ValueError("Domain mismatch")
            self._val = np.array(val.val,dtype=dtype,copy=copy)
        elif (np.isscalar(val)):
            self._val=np.full(global_shape,dtype=dtype,fill_value=val)
        elif isinstance(val, np.ndarray):
            if global_shape==val.shape:
                self._val = np.array(val,dtype=dtype,copy=copy)
            else:
                raise ValueError("Shape mismatch")
        elif val is None:
            self._val = np.empty(global_shape,dtype=dtype)
        else:
            raise TypeError("unknown source type")
csongor's avatar
csongor committed
107

108
    def _parse_domain(self, domain, val=None):
109
        if domain is None:
110
111
            if isinstance(val, Field):
                domain = val.domain
Martin Reinecke's avatar
Martin Reinecke committed
112
            elif np.isscalar(val):
113
                domain = ()
Martin Reinecke's avatar
Martin Reinecke committed
114
115
            else:
                raise TypeError("could not infer domain from value")
116
        elif isinstance(domain, DomainObject):
117
            domain = (domain,)
118
119
120
        elif not isinstance(domain, tuple):
            domain = tuple(domain)

csongor's avatar
csongor committed
121
        for d in domain:
122
            if not isinstance(d, DomainObject):
123
124
                raise TypeError(
                    "Given domain contains something that is not a "
125
                    "DomainObject instance.")
csongor's avatar
csongor committed
126
127
        return domain

128
129
    def _get_axes_tuple(self, things_with_shape):
        i = 0
Theo Steininger's avatar
Theo Steininger committed
130
131
        axes_list = []
        for thing in things_with_shape:
132
133
134
            nax = len(thing.shape)
            axes_list += [tuple(range(i,i+nax))]
            i += nax
Theo Steininger's avatar
Theo Steininger committed
135
        return tuple(axes_list)
136

137
    def _infer_dtype(self, dtype, val):
Martin Reinecke's avatar
cleanup    
Martin Reinecke committed
138
139
        if val is None:
            return np.float64 if dtype is None else dtype
csongor's avatar
csongor committed
140
        if dtype is None:
Martin Reinecke's avatar
cleanup    
Martin Reinecke committed
141
142
143
            if isinstance(val,Field):
                return val.dtype
            return np.result_type(val)
144

145
    # ---Factory methods---
146

147
    @classmethod
Martin Reinecke's avatar
stage1    
Martin Reinecke committed
148
    def from_random(cls, random_type, domain=None, dtype=None, **kwargs):
149
150
151
152
153
        """ Draws a random field with the given parameters.

        Parameters
        ----------
        cls : class
Theo Steininger's avatar
Theo Steininger committed
154

155
156
157
        random_type : String
            'pm1', 'normal', 'uniform' are the supported arguments for this
            method.
Theo Steininger's avatar
Theo Steininger committed
158

159
160
        domain : DomainObject
            The domain of the output random field
Theo Steininger's avatar
Theo Steininger committed
161

162
163
        dtype : type
            The datatype of the output random field
Theo Steininger's avatar
Theo Steininger committed
164

165
166
167
168
169
170
171
        Returns
        -------
        out : Field
            The output object.

        See Also
        --------
172
        power_synthesize
Theo Steininger's avatar
Theo Steininger committed
173

174
175

        """
Theo Steininger's avatar
Theo Steininger committed
176

Martin Reinecke's avatar
stage1    
Martin Reinecke committed
177
        f = cls(domain=domain, dtype=dtype)
178
        generator_function = getattr(Random, random_type)
Martin Reinecke's avatar
cleanup    
Martin Reinecke committed
179
        f.val=generator_function(dtype=f.dtype, shape=f.shape, **kwargs)
180
181
        return f

182
183
    # ---Powerspectral methods---

Martin Reinecke's avatar
Martin Reinecke committed
184
185
    def power_analyze(self, spaces=None, binbounds=None,
                      keep_phase_information=False):
Theo Steininger's avatar
Theo Steininger committed
186
        """ Computes the square root power spectrum for a subspace of `self`.
Theo Steininger's avatar
Theo Steininger committed
187

Theo Steininger's avatar
Theo Steininger committed
188
189
190
        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
191
        harmonic space. The resulting field has the same units as the initial
Theo Steininger's avatar
Theo Steininger committed
192
        field, corresponding to the square root of the power spectrum.
193
194
195

        Parameters
        ----------
Theo Steininger's avatar
Theo Steininger committed
196
197
198
199
200
        spaces : int *optional*
            The subspace for which the powerspectrum shall be computed
            (default : None).
        binbounds : array-like *optional*
            Inner bounds of the bins (default : None).
Martin Reinecke's avatar
Martin Reinecke committed
201
            if binbounds==None : bins are inferred.
202
203
204
205
206
207
208
209
210
211
        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
212

213
214
215
216
        Raise
        -----
        ValueError
            Raised if
Theo Steininger's avatar
Theo Steininger committed
217
218
                *len(domain) is != 1 when spaces==None
                *len(spaces) is != 1 if not None
219
                *the analyzed space is not harmonic
Theo Steininger's avatar
Theo Steininger committed
220

221
222
        Returns
        -------
Theo Steininger's avatar
Theo Steininger committed
223
        out : Field
Martin Reinecke's avatar
Martin Reinecke committed
224
            The output object. Its domain is a PowerSpace and it contains
225
226
227
228
229
            the power spectrum of 'self's field.

        See Also
        --------
        power_synthesize, PowerSpace
Theo Steininger's avatar
Theo Steininger committed
230

231
        """
Theo Steininger's avatar
Theo Steininger committed
232

Theo Steininger's avatar
Theo Steininger committed
233
        # check if all spaces in `self.domain` are either harmonic or
234
235
236
        # power_space instances
        for sp in self.domain:
            if not sp.harmonic and not isinstance(sp, PowerSpace):
Martin Reinecke's avatar
Martin Reinecke committed
237
                raise TypeError(
238
                    "Field has a space in `domain` which is neither "
239
240
241
                    "harmonic nor a PowerSpace.")

        # check if the `spaces` input is valid
242
243
        spaces = utilities.cast_axis_to_tuple(spaces, len(self.domain))
        if spaces is None:
Martin Reinecke's avatar
Martin Reinecke committed
244
            spaces = list(range(len(self.domain)))
245
246

        if len(spaces) == 0:
Martin Reinecke's avatar
cleanup    
Martin Reinecke committed
247
            raise ValueError("No space for analysis specified.")
248

249
250
251
252
253
254
255
256
257
258
        if keep_phase_information:
            parts_val = self._hermitian_decomposition(
                                              val=self.val,
                                              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]
259
260

        for space_index in spaces:
261
262
            parts = [self._single_power_analyze(
                                work_field=part,
263
                                space_index=space_index,
264
265
                                binbounds=binbounds)
                     for part in parts]
266

267
268
269
270
271
272
        if keep_phase_information:
            result_field = parts[0] + 1j*parts[1]
        else:
            result_field = parts[0]

        return result_field
273
274

    @classmethod
Martin Reinecke's avatar
Martin Reinecke committed
275
    def _single_power_analyze(cls, work_field, space_index, binbounds):
276

277
        if not work_field.domain[space_index].harmonic:
278
279
            raise ValueError(
                "The analyzed space must be harmonic.")
280

281
282
283
284
285
286
        # 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.

287
        harmonic_domain = work_field.domain[space_index]
288
        power_domain = PowerSpace(harmonic_partner=harmonic_domain,
Theo Steininger's avatar
Theo Steininger committed
289
                                  binbounds=binbounds)
290
291
        power_spectrum = cls._calculate_power_spectrum(
                                field_val=work_field.val,
Martin Reinecke's avatar
Martin Reinecke committed
292
                                pdomain=power_domain,
293
                                axes=work_field.domain_axes[space_index])
294
295

        # create the result field and put power_spectrum into it
296
        result_domain = list(work_field.domain)
297
        result_domain[space_index] = power_domain
298
        result_dtype = power_spectrum.dtype
299

300
        result_field = work_field.copy_empty(
301
                   domain=result_domain,
Martin Reinecke's avatar
stage1    
Martin Reinecke committed
302
                   dtype=result_dtype)
303
304
305
306
        result_field.set_val(new_val=power_spectrum, copy=False)

        return result_field

307
    @classmethod
Martin Reinecke's avatar
Martin Reinecke committed
308
    def _calculate_power_spectrum(cls, field_val, pdomain, axes=None):
309

Martin Reinecke's avatar
Martin Reinecke committed
310
311
312
        pindex = pdomain.pindex
        # MR FIXME: how about iterating over slices, instead of replicating
        # pindex? Would save memory and probably isn't slower.
313
        if axes is not None:
314
315
316
317
            pindex = cls._shape_up_pindex(
                            pindex=pindex,
                            target_shape=field_val.shape,
                            axes=axes)
Theo Steininger's avatar
Theo Steininger committed
318

Martin Reinecke's avatar
Martin Reinecke committed
319
        power_spectrum = utilities.bincount_axis(pindex, weights=field_val,
320
                                         axis=axes)
Martin Reinecke's avatar
Martin Reinecke committed
321
        rho = pdomain.rho
322
323
324
325
326
327
328
329
        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

330
    @staticmethod
Martin Reinecke's avatar
Martin Reinecke committed
331
    def _shape_up_pindex(pindex, target_shape, axes):
Martin Reinecke's avatar
cleanup    
Martin Reinecke committed
332
        semiscaled_local_shape = [1] * len(target_shape)
Theo Steininger's avatar
Theo Steininger committed
333
        for i in range(len(axes)):
Martin Reinecke's avatar
Martin Reinecke committed
334
            semiscaled_local_shape[axes[i]] = pindex.shape[i]
Martin Reinecke's avatar
cleanup    
Martin Reinecke committed
335
        semiscaled_local_data = pindex.reshape(semiscaled_local_shape)
Martin Reinecke's avatar
Martin Reinecke committed
336
        result_obj = np.empty(target_shape, dtype=pindex.dtype)
337
        result_obj[()] = semiscaled_local_data
338
339
        return result_obj

340
    def power_synthesize(self, spaces=None, real_power=True, real_signal=True,
Martin Reinecke's avatar
stage1    
Martin Reinecke committed
341
                         mean=None, std=None):
Theo Steininger's avatar
Theo Steininger committed
342
        """ Yields a sampled field with `self`**2 as its power spectrum.
Theo Steininger's avatar
Theo Steininger committed
343

Theo Steininger's avatar
Theo Steininger committed
344
        This method draws a Gaussian random field in the harmonic partner
Martin Reinecke's avatar
typos    
Martin Reinecke committed
345
        domain of this field's domains, using this field as power spectrum.
Theo Steininger's avatar
Theo Steininger committed
346

347
348
349
        Parameters
        ----------
        spaces : {tuple, int, None} *optional*
Theo Steininger's avatar
Theo Steininger committed
350
351
352
            Specifies the subspace containing all the PowerSpaces which
            should be converted (default : None).
            if spaces==None : Tries to convert the whole domain.
353
        real_power : boolean *optional*
Theo Steininger's avatar
Theo Steininger committed
354
355
            Determines whether the power spectrum is treated as intrinsically
            real or complex (default : True).
356
        real_signal : boolean *optional*
Theo Steininger's avatar
Theo Steininger committed
357
358
359
360
361
362
            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
363
        std : float *optional*
Theo Steininger's avatar
Theo Steininger committed
364
365
366
            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
367

368
369
370
371
        Returns
        -------
        out : Field
            The output object. A random field created with the power spectrum
Theo Steininger's avatar
Theo Steininger committed
372
            stored in the `spaces` in `self`.
373

Theo Steininger's avatar
Theo Steininger committed
374
375
376
377
378
379
        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.

380
381
382
        See Also
        --------
        power_analyze
Theo Steininger's avatar
Theo Steininger committed
383
384
385
386
387

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

388
        """
Theo Steininger's avatar
Theo Steininger committed
389

390
391
        # check if the `spaces` input is valid
        spaces = utilities.cast_axis_to_tuple(spaces, len(self.domain))
Theo Steininger's avatar
Theo Steininger committed
392
        if spaces is None:
Martin Reinecke's avatar
Martin Reinecke committed
393
            spaces = list(range(len(self.domain)))
Theo Steininger's avatar
Theo Steininger committed
394

395
396
397
398
399
        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.")
400
401
402

        # create the result domain
        result_domain = list(self.domain)
403
404
        for power_space_index in spaces:
            power_space = self.domain[power_space_index]
405
            harmonic_domain = power_space.harmonic_partner
406
            result_domain[power_space_index] = harmonic_domain
407
408
409

        # create random samples: one or two, depending on whether the
        # power spectrum is real or complex
410
411
        result_list = [self.__class__.from_random(
                             'normal',
412
413
414
                             mean=mean,
                             std=std,
                             domain=result_domain,
Martin Reinecke's avatar
stage1    
Martin Reinecke committed
415
                             dtype=np.complex)
Martin Reinecke's avatar
cleanup    
Martin Reinecke committed
416
                       for x in range(1 if real_power else 2)]
417
418
419
420
421

        # 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
Martin Reinecke's avatar
cleanup    
Martin Reinecke committed
422
        spec = np.sqrt(self.val)
423
        for power_space_index in spaces:
Martin Reinecke's avatar
cleanup    
Martin Reinecke committed
424
            spec = self._spec_to_rescaler(spec, power_space_index)
425
426

        # apply the rescaler to the random fields
Martin Reinecke's avatar
cleanup    
Martin Reinecke committed
427
        result_list[0].val *= spec.real
428
        if not real_power:
Martin Reinecke's avatar
cleanup    
Martin Reinecke committed
429
            result_list[1].val *= spec.imag
430

431
        if real_signal:
Martin Reinecke's avatar
cleanup    
Martin Reinecke committed
432
433
434
            for i in result_list:
                i.val = self._hermitian_decomposition(
                                            i.val,
435
                                            preserve_gaussian_variance=True)[0]
436
437
438

        if real_power:
            result = result_list[0]
Martin Reinecke's avatar
cleanup    
Martin Reinecke committed
439
            if not issubclass(result_list[0].dtype.type,
440
441
                              np.complexfloating):
                result = result.real
442
        else:
443
444
445
446
            result = result_list[0] + 1j*result_list[1]

        return result

447
    @staticmethod
Martin Reinecke's avatar
cleanup    
Martin Reinecke committed
448
    def _hermitian_decomposition(val, preserve_gaussian_variance=False):
449
        if preserve_gaussian_variance:
Martin Reinecke's avatar
cleanup    
Martin Reinecke committed
450
451
452
453
454
            if not issubclass(val.dtype.type, np.complexfloating):
                raise TypeError("complex input field is needed here")
            return (val.real*np.sqrt(2.), val.imag*np.sqrt(2.))
        else:
            return (val.real.copy(), val.imag.copy())
455

Martin Reinecke's avatar
cleanup    
Martin Reinecke committed
456
    def _spec_to_rescaler(self, spec, power_space_index):
457
        power_space = self.domain[power_space_index]
458

459
460
461
462
        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)
463
        local_blow_up[index] = power_space.pindex
464
        # here, the power_spectrum is distributed into the new shape
465
        return spec[local_blow_up]
466

Theo Steininger's avatar
Theo Steininger committed
467
    # ---Properties---
468

Theo Steininger's avatar
Theo Steininger committed
469
    def set_val(self, new_val=None, copy=False):
Martin Reinecke's avatar
stage1    
Martin Reinecke committed
470
        """ Sets the field's data object.
471
472
473

        Parameters
        ----------
Theo Steininger's avatar
Theo Steininger committed
474
        new_val : scalar, array-like, Field, None *optional*
475
476
            The values to be stored in the field.
            {default : None}
Theo Steininger's avatar
Theo Steininger committed
477

478
        copy : boolean, *optional*
Theo Steininger's avatar
Theo Steininger committed
479
480
            If False, Field tries to not copy the input data but use it
            directly.
481
482
483
484
485
486
            {default : False}
        See Also
        --------
        val

        """
Theo Steininger's avatar
Theo Steininger committed
487

Martin Reinecke's avatar
Martin Reinecke committed
488
489
490
491
492
493
494
495
        if new_val is None:
            pass
        elif isinstance(new_val, Field):
            if self.domain!=new_val.domain:
                raise ValueError("Domain mismatch")
            if copy:
                self._val[()] = new_val.val
            else:
Martin Reinecke's avatar
cleanup    
Martin Reinecke committed
496
                self._val = np.array(new_val.val,dtype=self.dtype,copy=False)
Martin Reinecke's avatar
Martin Reinecke committed
497
498
499
500
501
502
503
504
        elif (np.isscalar(new_val)):
            self._val[()]=new_val
        elif isinstance(new_val, np.ndarray):
            if copy:
                self._val[()] = new_val
            else:
                if self.shape!=new_val.shape:
                    raise ValueError("Shape mismatch")
Martin Reinecke's avatar
cleanup    
Martin Reinecke committed
505
                self._val = np.array(new_val,dtype=self.dtype,copy=False)
Martin Reinecke's avatar
Martin Reinecke committed
506
507
        else:
            raise TypeError("unknown source type")
508
        return self
csongor's avatar
csongor committed
509

510
    def get_val(self, copy=False):
Martin Reinecke's avatar
stage1    
Martin Reinecke committed
511
        """ Returns the data object associated with this Field.
512
513
514
515

        Parameters
        ----------
        copy : boolean
Martin Reinecke's avatar
stage1    
Martin Reinecke committed
516
            If true, a copy of the Field's underlying data object
Theo Steininger's avatar
Theo Steininger committed
517
            is returned.
Theo Steininger's avatar
Theo Steininger committed
518

519
520
        Returns
        -------
Martin Reinecke's avatar
stage1    
Martin Reinecke committed
521
        out : numpy.ndarray
522
523
524
525
526
527

        See Also
        --------
        val

        """
Theo Steininger's avatar
Theo Steininger committed
528

Martin Reinecke's avatar
Martin Reinecke committed
529
        return self._val.copy() if copy else self._val
csongor's avatar
csongor committed
530

Theo Steininger's avatar
Theo Steininger committed
531
532
    @property
    def val(self):
Martin Reinecke's avatar
stage1    
Martin Reinecke committed
533
        """ Returns the data object associated with this Field.
Theo Steininger's avatar
Theo Steininger committed
534

535
536
        Returns
        -------
Martin Reinecke's avatar
stage1    
Martin Reinecke committed
537
        out : numpy.ndarray
538
539
540
541
542
543

        See Also
        --------
        get_val

        """
Theo Steininger's avatar
Theo Steininger committed
544

Martin Reinecke's avatar
Martin Reinecke committed
545
        return self._val
csongor's avatar
csongor committed
546

Theo Steininger's avatar
Theo Steininger committed
547
548
    @val.setter
    def val(self, new_val):
549
        self.set_val(new_val=new_val, copy=False)
csongor's avatar
csongor committed
550

Martin Reinecke's avatar
Martin Reinecke committed
551
552
553
554
    @property
    def dtype(self):
        return self._val.dtype

555
556
    @property
    def shape(self):
Theo Steininger's avatar
Theo Steininger committed
557
        """ Returns the total shape of the Field's data array.
Theo Steininger's avatar
Theo Steininger committed
558

559
560
561
        Returns
        -------
        out : tuple
Martin Reinecke's avatar
Martin Reinecke committed
562
            The output object. The tuple contains the dimensions of the spaces
563
564
565
566
567
568
569
            in domain.

        See Also
        --------
        dim

        """
Martin Reinecke's avatar
Martin Reinecke committed
570
        return self._val.shape
csongor's avatar
csongor committed
571

572
573
    @property
    def dim(self):
Theo Steininger's avatar
Theo Steininger committed
574
        """ Returns the total number of pixel-dimensions the field has.
Theo Steininger's avatar
Theo Steininger committed
575

Theo Steininger's avatar
Theo Steininger committed
576
        Effectively, all values from shape are multiplied.
Theo Steininger's avatar
Theo Steininger committed
577

578
579
580
581
582
583
584
585
586
587
        Returns
        -------
        out : int
            The dimension of the Field.

        See Also
        --------
        shape

        """
Martin Reinecke's avatar
cleanup    
Martin Reinecke committed
588
        return self._val.size
csongor's avatar
csongor committed
589

590
591
    @property
    def dof(self):
Theo Steininger's avatar
Theo Steininger committed
592
593
594
595
596
597
        """ 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
598
599
600
601
602
603
604
        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
605
606
607
        """ Returns the total volume of all spaces in the domain.
        """

Theo Steininger's avatar
Theo Steininger committed
608
        volume_tuple = tuple(sp.total_volume for sp in self.domain)
609
        try:
Theo Steininger's avatar
Theo Steininger committed
610
            return reduce(lambda x, y: x * y, volume_tuple)
611
        except TypeError:
Theo Steininger's avatar
Theo Steininger committed
612
            return 0.
613

Theo Steininger's avatar
Theo Steininger committed
614
615
616
617
    @property
    def real(self):
        """ The real part of the field (data is not copied).
        """
618
        return Field(self.domain,self.val.real)
Theo Steininger's avatar
Theo Steininger committed
619
620
621
622
623

    @property
    def imag(self):
        """ The imaginary part of the field (data is not copied).
        """
624
        return Field(self.domain,self.val.imag)
Theo Steininger's avatar
Theo Steininger committed
625

Theo Steininger's avatar
Theo Steininger committed
626
    # ---Special unary/binary operations---
627

csongor's avatar
csongor committed
628

Martin Reinecke's avatar
stage1    
Martin Reinecke committed
629
    def copy(self, domain=None, dtype=None):
630
        """ Returns a full copy of the Field.
Theo Steininger's avatar
Theo Steininger committed
631

632
633
        If no keyword arguments are given, the returned object will be an
        identical copy of the original Field. By explicit specification one is
Martin Reinecke's avatar
stage1    
Martin Reinecke committed
634
        able to define the domain and the dtype of the returned Field.
635
636
637
638
639

        Parameters
        ----------
        domain : DomainObject
            The new domain the Field shall have.
Theo Steininger's avatar
Theo Steininger committed
640

641
642
        dtype : type
            The new dtype the Field shall have.
Theo Steininger's avatar
Theo Steininger committed
643

644
645
646
647
648
649
650
651
652
653
        Returns
        -------
        out : Field
            The output object. An identical copy of 'self'.

        See Also
        --------
        copy_empty

        """
Theo Steininger's avatar
Theo Steininger committed
654

655
656
657
        if domain is None:
            domain = self.domain
        return Field(domain=domain,val=self._val,dtype=dtype,copy=True)
csongor's avatar
csongor committed
658

Martin Reinecke's avatar
stage1    
Martin Reinecke committed
659
    def copy_empty(self, domain=None, dtype=None):
660
661
662
        """ 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
663
664
665
        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
Martin Reinecke's avatar
stage1    
Martin Reinecke committed
666
        to change the domain and the dtype of the returned Field.
Theo Steininger's avatar
Theo Steininger committed
667

668
669
670
671
        Parameters
        ----------
        domain : DomainObject
            The new domain the Field shall have.
Theo Steininger's avatar
Theo Steininger committed
672

673
674
        dtype : type
            The new dtype the Field shall have.
Theo Steininger's avatar
Theo Steininger committed
675

676
677
678
        Returns
        -------
        out : Field
Theo Steininger's avatar
Theo Steininger committed
679
            The output object.
680
681
682
683
684
685

        See Also
        --------
        copy

        """
Theo Steininger's avatar
Theo Steininger committed
686

Theo Steininger's avatar
Theo Steininger committed
687
688
689
690
        if domain is None:
            domain = self.domain
        if dtype is None:
            dtype = self.dtype
691
        return Field(domain=domain, dtype=dtype)
Theo Steininger's avatar
Theo Steininger committed
692
693

    def weight(self, power=1, inplace=False, spaces=None):
Theo Steininger's avatar
Theo Steininger committed
694
        """ Weights the pixels of `self` with their invidual pixel-volume.
695
696
697
698

        Parameters
        ----------
        power : number
Theo Steininger's avatar
Theo Steininger committed
699
            The pixels get weighted with the volume-factor**power.
Theo Steininger's avatar
Theo Steininger committed
700

701
        inplace : boolean
Theo Steininger's avatar
Theo Steininger committed
702
703
            If True, `self` will be weighted and returned. Otherwise, a copy
            is made.
Theo Steininger's avatar
Theo Steininger committed
704

Theo Steininger's avatar
Theo Steininger committed
705
706
        spaces : tuple of ints
            Determines on which subspace the operation takes place.
Theo Steininger's avatar
Theo Steininger committed
707

708
709
710
        Returns
        -------
        out : Field
Theo Steininger's avatar
Theo Steininger committed
711
            The weighted field.
712
713

        """
714
        new_field = self if inplace else self.copy_empty()
csongor's avatar
csongor committed
715

716
        new_val = self.get_val(copy=False)
csongor's avatar
csongor committed
717

718
        spaces = utilities.cast_axis_to_tuple(spaces, len(self.domain))
csongor's avatar
csongor committed
719
        if spaces is None:
Martin Reinecke's avatar
Martin Reinecke committed
720
            spaces = list(range(len(self.domain)))
csongor's avatar
csongor committed
721

722
        for ind, sp in enumerate(self.domain):
Theo Steininger's avatar
Theo Steininger committed
723
724
725
726
727
            if ind in spaces:
                new_val = sp.weight(new_val,
                                    power=power,
                                    axes=self.domain_axes[ind],
                                    inplace=inplace)
728
729

        new_field.set_val(new_val=new_val, copy=False)
csongor's avatar
csongor committed
730
731
        return new_field

Martin Reinecke's avatar
Martin Reinecke committed
732
    def vdot(self, x=None, spaces=None, bare=False):
Theo Steininger's avatar
Theo Steininger committed
733
        """ Computes the volume-factor-aware dot product of 'self' with x.
Theo Steininger's avatar
Theo Steininger committed
734

735
736
737
        Parameters
        ----------
        x : Field
Theo Steininger's avatar
Theo Steininger committed
738
            The domain of x must contain `self.domain`
Theo Steininger's avatar
Theo Steininger committed
739

Theo Steininger's avatar
Theo Steininger committed
740
741
742
        spaces : tuple of ints
            If the domain of `self` and `x` are not the same, `spaces` specfies
            the mapping.
Theo Steininger's avatar
Theo Steininger committed
743

744
        bare : boolean
Theo Steininger's avatar
Theo Steininger committed
745
            If true, no volume factors will be included in the computation.
Theo Steininger's avatar
Theo Steininger committed
746

747
748
749
        Returns
        -------
        out : float, complex
Theo Steininger's avatar
Theo Steininger committed
750

751
        """
752
753
754
        if not isinstance(x, Field):
            raise ValueError("The dot-partner must be an instance of " +
                             "the NIFTy field class")
Theo Steininger's avatar
Theo Steininger committed
755

Martin Reinecke's avatar
Martin Reinecke committed
756
        # Compute the dot respecting the fact of discrete/continuous spaces
757
        y = self if bare else self.weight(power=1)
Theo Steininger's avatar
Theo Steininger committed
758

759
        if spaces is None:
760
            return np.vdot(y.val.flatten(),x.val.flatten())
761
762
763
        else:
            # create a diagonal operator which is capable of taking care of the
            # axes-matching
764
            from .operators.diagonal_operator import DiagonalOperator
765
766
767
768
769
770
            diagonal = y.val.conjugate()
            diagonalOperator = DiagonalOperator(domain=y.domain,
                                                diagonal=diagonal,
                                                copy=False)
            dotted = diagonalOperator(x, spaces=spaces)
            return dotted.sum(spaces=spaces)
Theo Steininger's avatar
Theo Steininger committed
771

Theo Steininger's avatar
Theo Steininger committed
772
    def norm(self):
Martin Reinecke's avatar
tweaks    
Martin Reinecke committed
773
        """ Computes the L2-norm of the field values.
csongor's avatar
csongor committed
774

Theo Steininger's avatar
Theo Steininger committed
775
776
777
        Returns
        -------
        norm : scalar
Martin Reinecke's avatar
tweaks    
Martin Reinecke committed
778
            The L2-norm of the field values.
csongor's avatar
csongor committed
779
780

        """
Martin Reinecke's avatar
fixes    
Martin Reinecke committed
781
        return np.sqrt(np.abs(self.vdot(x=self)))
csongor's avatar
csongor committed
782
783

    def conjugate(self, inplace=False):
784
        """ Retruns the complex conjugate of the field.
Theo Steininger's avatar
Theo Steininger committed
785

786
787
788
        Parameters
        ----------
        inplace : boolean
Theo Steininger's avatar
Theo Steininger committed
789
            Decides whether the conjugation should be performed inplace.
Theo Steininger's avatar
Theo Steininger committed
790

791
792
793
794
        Returns
        -------
        cc : field
            The complex conjugated field.
csongor's avatar
csongor committed
795
796

        """
Theo Steininger's avatar
Theo Steininger committed
797

csongor's avatar
csongor committed
798
        if inplace:
799
800
            self.imag*=-1
            return self
csongor's avatar
csongor committed
801
        else:
802
            return Field(self.domain,np.conj(self.val),self.dtype)
csongor's avatar
csongor committed
803

Theo Steininger's avatar
Theo Steininger committed
804
    # ---General unary/contraction methods---
805

Theo Steininger's avatar
Theo Steininger committed
806
807
    def __pos__(self):
        return self.copy()
808

Theo Steininger's avatar
Theo Steininger committed
809
    def __neg__(self):
810
        return Field(self.domain,-self.val,self.dtype)
csongor's avatar
csongor committed
811

Theo Steininger's avatar
Theo Steininger committed
812
    def __abs__(self):
813
        return Field(self.domain,np.abs(self.val),self.dtype)
csongor's avatar
csongor committed
814

815
    def _contraction_helper(self, op, spaces):
Theo Steininger's avatar
Theo Steininger committed
816
        if spaces is None:
817
818
819
            return getattr(self.val, op)()
        # build a list of all axes
        spaces = utilities.cast_axis_to_tuple(spaces, len(self.domain))
csongor's avatar
csongor committed
820

821
        axes_list = tuple(self.domain_axes[sp_index] for sp_index in spaces)
822
823

        try:
Theo Steininger's avatar
Theo Steininger committed
824
            axes_list = reduce(lambda x, y: x+y, axes_list)
825
        except TypeError:
Theo Steininger's avatar
Theo Steininger committed
826
            axes_list = ()
csongor's avatar
csongor committed
827

Martin Reinecke's avatar
stage1    
Martin Reinecke committed
828
        # perform the contraction on the data
Theo Steininger's avatar
Theo Steininger committed
829
830
        data = self.get_val(copy=False)
        data = getattr(data, op)(axis=axes_list)
csongor's avatar
csongor committed
831

Theo Steininger's avatar
Theo Steininger committed
832
833
834
        # check if the result is scalar or if a result_field must be constr.
        if np.isscalar(data):
            return data
csongor's avatar
csongor committed
835
        else:
Theo Steininger's avatar
Theo Steininger committed
836
            return_domain = tuple(self.domain[i]
Martin Reinecke's avatar
Martin Reinecke committed
837
                                  for i in range(len(self.domain))
Theo Steininger's avatar
Theo Steininger committed
838
                                  if i not in spaces)
839

Theo Steininger's avatar
Theo Steininger committed
840
841
842
843
            return_field = Field(domain=return_domain,
                                 val=data,
                                 copy=False)
            return return_field
csongor's avatar
csongor committed
844

845
846
    def sum(self, spaces=None):
        return self._contraction_helper('sum', spaces)
csongor's avatar
csongor committed
847

848
849
    def prod(self, spaces=None):
        return self._contraction_helper('prod', spaces)
csongor's avatar
csongor committed
850

851
852
    def all(self, spaces=None):
        return self._contraction_helper('all', spaces)
csongor's avatar
csongor committed
853

854
855
    def any(self, spaces=None):
        return self._contraction_helper('any', spaces)
csongor's avatar
csongor committed
856

857
858
    def min(self, spaces=None):
        return self._contraction_helper('min', spaces)
csongor's avatar
csongor committed
859

860
861
    def nanmin(self, spaces=None):
        return self._contraction_helper('nanmin', spaces)
csongor's avatar
csongor committed
862

863
864
    def max(self, spaces=None):
        return self._contraction_helper('max', spaces)
csongor's avatar
csongor committed
865

866
867
    def nanmax(self, spaces=None):
        return self._contraction_helper('nanmax', spaces)
csongor's avatar
csongor committed
868

869
870
    def mean(self, spaces=None):
        return self._contraction_helper('mean', spaces)
csongor's avatar
csongor committed
871

872
873
    def var(self, spaces=None):
        return self._contraction_helper('var', spaces)
csongor's avatar
csongor committed
874

875
876
    def std(self, spaces=None):
        return self._contraction_helper('std', spaces)
csongor's avatar
csongor committed
877

Theo Steininger's avatar
Theo Steininger committed
878
    # ---General binary methods---
csongor's avatar
csongor committed
879

880
    def _binary_helper(self, other, op):
csongor's avatar
csongor committed
881
        # if other is a field, make sure that the domains match
882
        if isinstance(other, Field):
883
884
885
            if other.domain != self.domain:
                raise ValueError("domains are incompatible.")
            return Field(self.domain,getattr(self.val,op)(other.val))
csongor's avatar