test_power.py 10.5 KB
Newer Older
Philipp Arras's avatar
Philipp Arras committed
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
# 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/>.
#
# 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.

import unittest
import nifty4 as ift
import numpy as np
from itertools import product
from test.common import expand
from numpy.testing import assert_allclose


# TODO Add also other space types


30
class Energy_Tests(unittest.TestCase):
Philipp Arras's avatar
Philipp Arras committed
31
    @expand(product([ift.RGSpace(64, distances=.789),
32
33
34
35
                     ift.RGSpace([32, 32], distances=.789)],
                    [132, 42, 3]))
    def testLinearPower(self, space, seed):
        np.random.seed(seed)
Philipp Arras's avatar
Philipp Arras committed
36
37
38
39
40
41
42
43
        dim = len(space.shape)
        hspace = space.get_default_codomain()
        ht = ift.HarmonicTransformOperator(hspace, space)
        binbounds = ift.PowerSpace.useful_binbounds(hspace, logarithmic=True)
        pspace = ift.PowerSpace(hspace, binbounds=binbounds)
        P = ift.PowerProjectionOperator(domain=hspace, power_space=pspace)
        xi = ift.Field.from_random(domain=hspace, random_type='normal')

Martin Reinecke's avatar
Martin Reinecke committed
44
        # TODO Power spectrum should depend on number of pixels
45
46
47
48
49
50
        def pspec(k): return 64 / (1 + k**2)**dim
        pspec = ift.PS_field(pspace, pspec)
        tau0 = ift.log(pspec)
        A = P.adjoint_times(ift.sqrt(pspec))
        n = ift.Field.from_random(domain=space, random_type='normal', std=.01)
        N = ift.DiagonalOperator(n**2)
Philipp Arras's avatar
Philipp Arras committed
51
        s = xi * A
Martin Reinecke's avatar
Martin Reinecke committed
52
        Instrument = ift.ScalingOperator(1., space)
Philipp Arras's avatar
Philipp Arras committed
53
54
55
56
57
        R = Instrument * ht
        d = R(s) + n

        direction = ift.Field.from_random('normal', pspace)
        direction /= np.sqrt(direction.var())
58
        eps = 1e-7
Philipp Arras's avatar
Philipp Arras committed
59
60
61
62
63
64
65
66
67
68
69
70
        tau1 = tau0 + eps * direction

        IC = ift.GradientNormController(
            iteration_limit=100,
            tol_abs_gradnorm=1e-5)
        inverter = ift.ConjugateGradient(IC)

        S = ift.create_power_operator(hspace, power_spectrum=lambda k: 1.)

        D = ift.library.WienerFilterEnergy(position=s, d=d, R=R, N=N, S=S,
                                           inverter=inverter).curvature

71
72
73
74
75
76
77
        w = ift.Field.zeros_like(tau0)
        Nsamples = 10
        for i in range(Nsamples):
            sample = D.generate_posterior_sample() + s
            w += P(abs(sample)**2)
        w /= Nsamples

Philipp Arras's avatar
Philipp Arras committed
78
        energy0 = ift.library.CriticalPowerEnergy(
79
            position=tau0, m=s, inverter=inverter, w=w)
Philipp Arras's avatar
Philipp Arras committed
80
        energy1 = ift.library.CriticalPowerEnergy(
81
            position=tau1, m=s, inverter=inverter, w=w)
Philipp Arras's avatar
Philipp Arras committed
82
83
84

        a = (energy1.value - energy0.value) / eps
        b = energy0.gradient.vdot(direction)
85
        tol = 1e-4
Philipp Arras's avatar
Philipp Arras committed
86
87
88
89
        assert_allclose(a, b, rtol=tol, atol=tol)

    @expand(product([ift.RGSpace(64, distances=.789),
                     ift.RGSpace([32, 32], distances=.789)],
90
91
92
93
                    [ift.library.Exponential, ift.library.Linear],
                    [132, 42, 3]))
    def testNonlinearPower(self, space, nonlinearity, seed):
        np.random.seed(seed)
Philipp Arras's avatar
Philipp Arras committed
94
95
        f = nonlinearity()
        dim = len(space.shape)
96
97
        hspace = space.get_default_codomain()
        ht = ift.HarmonicTransformOperator(hspace, space)
Philipp Arras's avatar
Philipp Arras committed
98
99
100
101
102
103
104
105
106
        binbounds = ift.PowerSpace.useful_binbounds(hspace, logarithmic=True)
        pspace = ift.PowerSpace(hspace, binbounds=binbounds)
        P = ift.PowerProjectionOperator(domain=hspace, power_space=pspace)
        xi = ift.Field.from_random(domain=hspace, random_type='normal')

        def pspec(k): return 1 / (1 + k**2)**dim
        tau0 = ift.PS_field(pspace, pspec)
        A = P.adjoint_times(ift.sqrt(tau0))
        n = ift.Field.from_random(domain=space, random_type='normal')
107
        s = ht(xi * A)
Martin Reinecke's avatar
Martin Reinecke committed
108
        R = ift.ScalingOperator(10., space)
Philipp Arras's avatar
Philipp Arras committed
109
110
111
112
113
114
        diag = ift.Field.ones(space)
        N = ift.DiagonalOperator(diag)
        d = R(f(s)) + n

        direction = ift.Field.from_random('normal', pspace)
        direction /= np.sqrt(direction.var())
115
        eps = 1e-7
Philipp Arras's avatar
Philipp Arras committed
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
        tau1 = tau0 + eps * direction

        IC = ift.GradientNormController(
            iteration_limit=100,
            tol_abs_gradnorm=1e-5)
        inverter = ift.ConjugateGradient(IC)

        S = ift.create_power_operator(hspace, power_spectrum=lambda k: 1.)
        D = ift.library.NonlinearWienerFilterEnergy(
            position=xi,
            d=d,
            Instrument=R,
            nonlinearity=f,
            power=A,
            N=N,
            S=S,
132
            ht=ht,
Philipp Arras's avatar
Philipp Arras committed
133
            inverter=inverter).curvature
134
        Nsamples = 10
Philipp Arras's avatar
Philipp Arras committed
135
136
137
        xi_sample_list = [
            D.generate_posterior_sample() +
            xi for _ in range(Nsamples)]
Philipp Arras's avatar
Philipp Arras committed
138
139
140
141

        energy0 = ift.library.NonlinearPowerEnergy(
            position=tau0,
            d=d,
Philipp Arras's avatar
Philipp Arras committed
142
            xi=xi,
Philipp Arras's avatar
Philipp Arras committed
143
144
145
146
            D=D,
            Instrument=R,
            Projection=P,
            nonlinearity=f,
147
            ht=ht,
Philipp Arras's avatar
Philipp Arras committed
148
            N=N,
Philipp Arras's avatar
Philipp Arras committed
149
            xi_sample_list=xi_sample_list)
Philipp Arras's avatar
Philipp Arras committed
150
151
152
        energy1 = ift.library.NonlinearPowerEnergy(
            position=tau1,
            d=d,
Philipp Arras's avatar
Philipp Arras committed
153
            xi=xi,
Philipp Arras's avatar
Philipp Arras committed
154
155
156
157
            D=D,
            Instrument=R,
            Projection=P,
            nonlinearity=f,
158
            ht=ht,
Philipp Arras's avatar
Philipp Arras committed
159
            N=N,
Philipp Arras's avatar
Philipp Arras committed
160
            xi_sample_list=xi_sample_list)
Philipp Arras's avatar
Philipp Arras committed
161
162
163

        a = (energy1.value - energy0.value) / eps
        b = energy0.gradient.vdot(direction)
164
        tol = 1e-4
Philipp Arras's avatar
Philipp Arras committed
165
        assert_allclose(a, b, rtol=tol, atol=tol)
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181


class Curvature_Tests(unittest.TestCase):
    @expand(product([ift.RGSpace(64, distances=.789),
                     ift.RGSpace([32, 32], distances=.789)],
                    [132, 42, 3]))
    def testLinearPowerCurvature(self, space, seed):
        np.random.seed(seed)
        dim = len(space.shape)
        hspace = space.get_default_codomain()
        ht = ift.HarmonicTransformOperator(hspace, space)
        binbounds = ift.PowerSpace.useful_binbounds(hspace, logarithmic=True)
        pspace = ift.PowerSpace(hspace, binbounds=binbounds)
        P = ift.PowerProjectionOperator(domain=hspace, power_space=pspace)
        xi = ift.Field.from_random(domain=hspace, random_type='normal')

Martin Reinecke's avatar
Martin Reinecke committed
182
        # TODO Power spectrum should depend on number of pixels
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
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
269
270
271
272
273
274
275
        def pspec(k): return 64 / (1 + k**2)**dim
        pspec = ift.PS_field(pspace, pspec)
        tau0 = ift.log(pspec)
        A = P.adjoint_times(ift.sqrt(pspec))
        n = ift.Field.from_random(domain=space, random_type='normal', std=.01)
        N = ift.DiagonalOperator(n**2)
        s = xi * A
        diag = ift.Field.ones(space)
        Instrument = ift.DiagonalOperator(diag)
        R = Instrument * ht
        d = R(s) + n

        direction = ift.Field.from_random('normal', pspace)
        direction /= np.sqrt(direction.var())
        eps = 1e-7
        tau1 = tau0 + eps * direction

        IC = ift.GradientNormController(
            iteration_limit=100,
            tol_abs_gradnorm=1e-5)
        inverter = ift.ConjugateGradient(IC)

        S = ift.create_power_operator(hspace, power_spectrum=lambda k: 1.)

        D = ift.library.WienerFilterEnergy(position=s, d=d, R=R, N=N, S=S,
                                           inverter=inverter).curvature

        w = ift.Field.zeros_like(tau0)
        Nsamples = 10
        for i in range(Nsamples):
            sample = D.generate_posterior_sample() + s
            w += P(abs(sample)**2)
        w /= Nsamples

        energy0 = ift.library.CriticalPowerEnergy(
            position=tau0, m=s, inverter=inverter, w=w)

        gradient0 = energy0.gradient
        gradient1 = energy0.at(tau1).gradient

        a = (gradient1 - gradient0) / eps
        b = energy0.curvature(direction)
        tol = 1e-5
        assert_allclose(a.val, b.val, rtol=tol, atol=tol)

    @expand(product([ift.RGSpace(64, distances=.789),
                     ift.RGSpace([32, 32], distances=.789)],
                    [ift.library.Exponential, ift.library.Linear],
                    [132, 42, 3]))
    def testNonlinearPowerCurvature(self, space, nonlinearity, seed):
        np.random.seed(seed)
        f = nonlinearity()
        dim = len(space.shape)
        hspace = space.get_default_codomain()
        ht = ift.HarmonicTransformOperator(hspace, space)
        binbounds = ift.PowerSpace.useful_binbounds(hspace, logarithmic=True)
        pspace = ift.PowerSpace(hspace, binbounds=binbounds)
        P = ift.PowerProjectionOperator(domain=hspace, power_space=pspace)
        xi = ift.Field.from_random(domain=hspace, random_type='normal')

        def pspec(k): return 1 / (1 + k**2)**dim
        tau0 = ift.PS_field(pspace, pspec)
        A = P.adjoint_times(ift.sqrt(tau0))
        n = ift.Field.from_random(domain=space, random_type='normal')
        s = ht(xi * A)
        diag = ift.Field.ones(space) * 10
        R = ift.DiagonalOperator(diag)
        diag = ift.Field.ones(space)
        N = ift.DiagonalOperator(diag)
        d = R(f(s)) + n

        direction = ift.Field.from_random('normal', pspace)
        direction /= np.sqrt(direction.var())
        eps = 1e-7
        tau1 = tau0 + eps * direction

        IC = ift.GradientNormController(
            iteration_limit=100,
            tol_abs_gradnorm=1e-5)
        inverter = ift.ConjugateGradient(IC)

        S = ift.create_power_operator(hspace, power_spectrum=lambda k: 1.)
        D = ift.library.NonlinearWienerFilterEnergy(
            position=xi,
            d=d,
            Instrument=R,
            nonlinearity=f,
            power=A,
            N=N,
            S=S,
            ht=ht,
            inverter=inverter).curvature
        Nsamples = 10
Philipp Arras's avatar
Philipp Arras committed
276
277
278
        xi_sample_list = [
            D.generate_posterior_sample() +
            xi for _ in range(Nsamples)]
279
280
281
282

        energy0 = ift.library.NonlinearPowerEnergy(
            position=tau0,
            d=d,
Philipp Arras's avatar
Philipp Arras committed
283
            xi=xi,
284
285
286
287
288
289
            D=D,
            Instrument=R,
            Projection=P,
            nonlinearity=f,
            ht=ht,
            N=N,
Philipp Arras's avatar
Philipp Arras committed
290
            xi_sample_list=xi_sample_list)
291
292
293
294
295
296

        gradient0 = energy0.gradient
        gradient1 = energy0.at(tau1).gradient

        a = (gradient1 - gradient0) / eps
        b = energy0.curvature(direction)
Philipp Arras's avatar
Philipp Arras committed
297
        tol = 1e-3
298
        assert_allclose(a.val, b.val, rtol=tol, atol=tol)