space.py 9.51 KB
Newer Older
1
2
# NIFTY (Numerical Information Field Theory) has been developed at the
# Max-Planck-Institute for Astrophysics.
Marco Selig's avatar
Marco Selig committed
3
##
4
# Copyright (C) 2013 Max-Planck-Society
Marco Selig's avatar
Marco Selig committed
5
##
6
7
# Author: Marco Selig
# Project homepage: <http://www.mpa-garching.mpg.de/ift/nifty/>
Marco Selig's avatar
Marco Selig committed
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.
Marco Selig's avatar
Marco Selig committed
13
##
14
15
16
17
# 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.
Marco Selig's avatar
Marco Selig committed
18
##
19
20
# You should have received a copy of the GNU General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
Marco Selig's avatar
Marco Selig committed
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44

"""
    ..                  __   ____   __
    ..                /__/ /   _/ /  /_
    ..      __ ___    __  /  /_  /   _/  __   __
    ..    /   _   | /  / /   _/ /  /   /  / /  /
    ..   /  / /  / /  / /  /   /  /_  /  /_/  /
    ..  /__/ /__/ /__/ /__/    \___/  \___   /  core
    ..                               /______/

    .. The NIFTY project homepage is http://www.mpa-garching.mpg.de/ift/nifty/

    NIFTY [#]_, "Numerical Information Field Theory", is a versatile
    library designed to enable the development of signal inference algorithms
    that operate regardless of the underlying spatial grid and its resolution.
    Its object-oriented framework is written in Python, although it accesses
    libraries written in Cython, C++, and C for efficiency.

    NIFTY offers a toolkit that abstracts discretized representations of
    continuous spaces, fields in these spaces, and operators acting on fields
    into classes. Thereby, the correct normalization of operations on fields is
    taken care of automatically without concerning the user. This allows for an
    abstract formulation and programming of inference algorithms, including
    those derived within information field theory. Thus, NIFTY permits its user
Marco Selig's avatar
Marco Selig committed
45
    to rapidly prototype algorithms in 1D and then apply the developed code in
Marco Selig's avatar
Marco Selig committed
46
47
48
49
50
    higher-dimensional settings of real world problems. The set of spaces on
    which NIFTY operates comprises point sets, n-dimensional regular grids,
    spherical spaces, their harmonic counterparts, and product spaces
    constructed as combinations of those.

51
52
53
54
55
56
57
    References
    ----------
    .. [#] Selig et al., "NIFTY -- Numerical Information Field Theory --
        a versatile Python library for signal inference",
        `A&A, vol. 554, id. A26 <http://dx.doi.org/10.1051/0004-6361/201321236>`_,
        2013; `arXiv:1301.4499 <http://www.arxiv.org/abs/1301.4499>`_

Marco Selig's avatar
Marco Selig committed
58
59
60
61
62
63
    Class & Feature Overview
    ------------------------
    The NIFTY library features three main classes: **spaces** that represent
    certain grids, **fields** that are defined on spaces, and **operators**
    that apply to fields.

64
65
    .. Overview of all (core) classes:
    ..
Marco Selig's avatar
Marco Selig committed
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
    .. - switch
    .. - notification
    .. - _about
    .. - random
    .. - space
    ..     - point_space
    ..     - rg_space
    ..     - lm_space
    ..     - gl_space
    ..     - hp_space
    ..     - nested_space
    .. - field
    .. - operator
    ..     - diagonal_operator
    ..         - power_operator
    ..     - projection_operator
    ..     - vecvec_operator
    ..     - response_operator
    .. - probing
    ..     - trace_probing
    ..     - diagonal_probing

88
89
    Overview of the main classes and functions:

Marco Selig's avatar
Marco Selig committed
90
91
    .. automodule:: nifty

92
93
94
95
96
97
98
99
100
101
102
103
104
105
    - :py:class:`space`
        - :py:class:`point_space`
        - :py:class:`rg_space`
        - :py:class:`lm_space`
        - :py:class:`gl_space`
        - :py:class:`hp_space`
        - :py:class:`nested_space`
    - :py:class:`field`
    - :py:class:`operator`
        - :py:class:`diagonal_operator`
            - :py:class:`power_operator`
        - :py:class:`projection_operator`
        - :py:class:`vecvec_operator`
        - :py:class:`response_operator`
Marco Selig's avatar
Marco Selig committed
106

107
        .. currentmodule:: nifty.nifty_tools
Marco Selig's avatar
Marco Selig committed
108

109
110
        - :py:class:`invertible_operator`
        - :py:class:`propagator_operator`
Marco Selig's avatar
Marco Selig committed
111

112
        .. currentmodule:: nifty.nifty_explicit
Marco Selig's avatar
Marco Selig committed
113

114
        - :py:class:`explicit_operator`
Marco Selig's avatar
Marco Selig committed
115

116
    .. automodule:: nifty
Marco Selig's avatar
Marco Selig committed
117

118
119
120
    - :py:class:`probing`
        - :py:class:`trace_probing`
        - :py:class:`diagonal_probing`
Marco Selig's avatar
Marco Selig committed
121

122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
        .. currentmodule:: nifty.nifty_explicit

        - :py:class:`explicit_probing`

    .. currentmodule:: nifty.nifty_tools

    - :py:class:`conjugate_gradient`
    - :py:class:`steepest_descent`

    .. currentmodule:: nifty.nifty_explicit

    - :py:func:`explicify`

    .. currentmodule:: nifty.nifty_power

    - :py:func:`weight_power`,
      :py:func:`smooth_power`,
      :py:func:`infer_power`,
      :py:func:`interpolate_power`
Marco Selig's avatar
Marco Selig committed
141
142
143

"""
from __future__ import division
144
145
146

import abc

Marco Selig's avatar
Marco Selig committed
147
148
import numpy as np

Theo Steininger's avatar
Theo Steininger committed
149
150
from keepers import Loggable,\
                    Versionable
Theo Steininger's avatar
Theo Steininger committed
151

152

153
class Space(Versionable, Loggable, object):
Marco Selig's avatar
Marco Selig committed
154
    """
Theo Steininger's avatar
Theo Steininger committed
155
156
157
158
159
160
161
        ..                            __             __
        ..                          /__/           /  /_
        ..      ______    ______    __   __ ___   /   _/
        ..    /   _   | /   _   | /  / /   _   | /  /
        ..   /  /_/  / /  /_/  / /  / /  / /  / /  /_
        ..  /   ____/  \______/ /__/ /__/ /__/  \___/  space class
        .. /__/
Marco Selig's avatar
Marco Selig committed
162

Theo Steininger's avatar
Theo Steininger committed
163
        NIFTY subclass for unstructured spaces.
Marco Selig's avatar
Marco Selig committed
164

Theo Steininger's avatar
Theo Steininger committed
165
166
        Unstructured spaces are lists of values without any geometrical
        information.
Marco Selig's avatar
Marco Selig committed
167
168
169

        Parameters
        ----------
Theo Steininger's avatar
Theo Steininger committed
170
171
        num : int
            Number of points.
172
        dtype : numpy.dtype, *optional*
Theo Steininger's avatar
Theo Steininger committed
173
            Data type of the field values (default: None).
Marco Selig's avatar
Marco Selig committed
174

Theo Steininger's avatar
Theo Steininger committed
175
        Attributes
Marco Selig's avatar
Marco Selig committed
176
        ----------
Theo Steininger's avatar
Theo Steininger committed
177
178
        para : numpy.ndarray
            Array containing the number of points.
179
        dtype : numpy.dtype
Theo Steininger's avatar
Theo Steininger committed
180
181
182
183
184
185
            Data type of the field values.
        discrete : bool
            Parameter captioning the fact that a :py:class:`point_space` is
            always discrete.
        vol : numpy.ndarray
            Pixel volume of the :py:class:`point_space`, which is always 1.
Marco Selig's avatar
Marco Selig committed
186
    """
187

188
189
190
    __metaclass__ = abc.ABCMeta

    def __init__(self, dtype=np.dtype('float')):
Theo Steininger's avatar
Theo Steininger committed
191
192
        """
            Sets the attributes for a point_space class instance.
Marco Selig's avatar
Marco Selig committed
193

Theo Steininger's avatar
Theo Steininger committed
194
195
196
197
            Parameters
            ----------
            num : int
                Number of points.
198
            dtype : numpy.dtype, *optional*
Theo Steininger's avatar
Theo Steininger committed
199
                Data type of the field values (default: numpy.float64).
Marco Selig's avatar
Marco Selig committed
200

Theo Steininger's avatar
Theo Steininger committed
201
202
203
204
            Returns
            -------
            None.
        """
205

206
        # parse dtype
207
208
209
210
        casted_dtype = np.result_type(dtype, np.float64)
        if casted_dtype != dtype:
            self.Logger.warning("Input dtype reset to: %s" % str(casted_dtype))
        self.dtype = casted_dtype
211

Theo Steininger's avatar
Theo Steininger committed
212
        self._ignore_for_hash = ['_global_id']
213

214
215
216
    def __hash__(self):
        # Extract the identifying parts from the vars(self) dict.
        result_hash = 0
Theo Steininger's avatar
Theo Steininger committed
217
218
        for key in sorted(vars(self).keys()):
            item = vars(self)[key]
219
            if key in self._ignore_for_hash or key == '_ignore_for_hash':
220
                continue
221
            result_hash ^= item.__hash__() ^ int(hash(key)/117)
222
223
        return result_hash

224
225
226
227
228
    def __eq__(self, x):
        if isinstance(x, type(self)):
            return hash(self) == hash(x)
        else:
            return False
229

230
231
232
    def __ne__(self, x):
        return not self.__eq__(x)

233
234
235
    @abc.abstractproperty
    def harmonic(self):
        raise NotImplementedError
236

237
    @abc.abstractproperty
238
    def shape(self):
239
240
        raise NotImplementedError(
            "There is no generic shape for the Space base class.")
Marco Selig's avatar
Marco Selig committed
241

242
    @abc.abstractproperty
243
    def dim(self):
244
245
        raise NotImplementedError(
            "There is no generic dim for the Space base class.")
Marco Selig's avatar
Marco Selig committed
246

247
    @abc.abstractproperty
248
    def total_volume(self):
249
250
        raise NotImplementedError(
            "There is no generic volume for the Space base class.")
251

252
253
254
    @abc.abstractmethod
    def copy(self):
        return self.__class__(dtype=self.dtype)
255

256
    @abc.abstractmethod
257
    def weight(self, x, power=1, axes=None, inplace=False):
Marco Selig's avatar
Marco Selig committed
258
        """
Theo Steininger's avatar
Theo Steininger committed
259
260
            Weights a given array of field values with the pixel volumes (not
            the meta volumes) to a given power.
Marco Selig's avatar
Marco Selig committed
261
262
263

            Parameters
            ----------
Theo Steininger's avatar
Theo Steininger committed
264
265
266
267
            x : numpy.ndarray
                Array to be weighted.
            power : float, *optional*
                Power of the pixel volumes to be used (default: 1).
Marco Selig's avatar
Marco Selig committed
268
269

            Returns
Theo Steininger's avatar
Theo Steininger committed
270
271
272
            -------
            y : numpy.ndarray
                Weighted array.
Marco Selig's avatar
Marco Selig committed
273
        """
274
        raise NotImplementedError
Theo Steininger's avatar
Theo Steininger committed
275

276
277
278
279
    def pre_cast(self, x, axes=None):
        return x

    def post_cast(self, x, axes=None):
280
281
        return x

282
    def get_distance_array(self, distribution_strategy):
283
        raise NotImplementedError(
284
285
            "There is no generic distance structure for Space base class.")

286
    def get_fft_smoothing_kernel_function(self, sigma):
287
288
        raise NotImplementedError(
            "There is no generic co-smoothing kernel for Space base class.")
289

290
291
292
    def hermitian_decomposition(self, x, axes=None):
        raise NotImplementedError

293
    def __repr__(self):
Theo Steininger's avatar
Theo Steininger committed
294
295
        string = ""
        string += str(type(self)) + "\n"
296
        string += "dtype: " + str(self.dtype) + "\n"
Theo Steininger's avatar
Theo Steininger committed
297
        return string
Theo Steininger's avatar
Theo Steininger committed
298
299
300
301
302
303
304
305
306
307
308
309

    # ---Serialization---

    def _to_hdf5(self, hdf5_group):
        hdf5_group.attrs['dtype'] = self.dtype.name

        return None

    @classmethod
    def _from_hdf5(cls, hdf5_group, repository):
        result = cls(dtype=np.dtype(hdf5_group.attrs['dtype']))
        return result