extra.py 4.98 KB
Newer Older
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
# 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) 2013-2018 Max-Planck-Society
#
# NIFTy is being developed at the Max-Planck-Institut fuer Astrophysik
# and financially supported by the Studienstiftung des deutschen Volkes.

import numpy as np
Philipp Arras's avatar
Philipp Arras committed
20

Martin Reinecke's avatar
fix  
Martin Reinecke committed
21
22
23
24
from .compat import *
from .field import Field
from .linearization import Linearization
from .sugar import from_random
25

Martin Reinecke's avatar
Martin Reinecke committed
26
__all__ = ["consistency_check", "check_value_gradient_consistency",
Martin Reinecke's avatar
Martin Reinecke committed
27
           "check_value_gradient_metric_consistency"]
28

Philipp Arras's avatar
Philipp Arras committed
29

Martin Reinecke's avatar
Martin Reinecke committed
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
def _assert_allclose(f1, f2, atol, rtol):
    if isinstance(f1, Field):
        return np.testing.assert_allclose(f1.local_data, f2.local_data,
                                          atol=atol, rtol=rtol)
    for key, val in f1.items():
        _assert_allclose(val, f2[key], atol=atol, rtol=rtol)


def _adjoint_implementation(op, domain_dtype, target_dtype, atol, rtol):
    needed_cap = op.TIMES | op.ADJOINT_TIMES
    if (op.capability & needed_cap) != needed_cap:
        return
    f1 = from_random("normal", op.domain, dtype=domain_dtype)
    f2 = from_random("normal", op.target, dtype=target_dtype)
    res1 = f1.vdot(op.adjoint_times(f2))
    res2 = op.times(f1).vdot(f2)
    np.testing.assert_allclose(res1, res2, atol=atol, rtol=rtol)


def _inverse_implementation(op, domain_dtype, target_dtype, atol, rtol):
    needed_cap = op.TIMES | op.INVERSE_TIMES
    if (op.capability & needed_cap) != needed_cap:
        return
    foo = from_random("normal", op.target, dtype=target_dtype)
    res = op(op.inverse_times(foo))
    _assert_allclose(res, foo, atol=atol, rtol=rtol)

    foo = from_random("normal", op.domain, dtype=domain_dtype)
    res = op.inverse_times(op(foo))
    _assert_allclose(res, foo, atol=atol, rtol=rtol)


def _full_implementation(op, domain_dtype, target_dtype, atol, rtol):
    _adjoint_implementation(op, domain_dtype, target_dtype, atol, rtol)
    _inverse_implementation(op, domain_dtype, target_dtype, atol, rtol)


def consistency_check(op, domain_dtype=np.float64, target_dtype=np.float64,
                      atol=0, rtol=1e-7):
    _full_implementation(op, domain_dtype, target_dtype, atol, rtol)
    _full_implementation(op.adjoint, target_dtype, domain_dtype, atol, rtol)
    _full_implementation(op.inverse, target_dtype, domain_dtype, atol, rtol)
    _full_implementation(op.adjoint.inverse, domain_dtype, target_dtype, atol,
                         rtol)


Martin Reinecke's avatar
Martin Reinecke committed
76
def _get_acceptable_location(op, loc, lin):
Martin Reinecke's avatar
Martin Reinecke committed
77
    if not np.isfinite(lin.val.sum()):
Martin Reinecke's avatar
Martin Reinecke committed
78
79
80
81
        raise ValueError('Initial value must be finite')
    dir = from_random("normal", loc.domain)
    dirder = lin.jac(dir)
    if dirder.norm() == 0:
Martin Reinecke's avatar
Martin Reinecke committed
82
        dir = dir * (lin.val.norm()*1e-5)
Martin Reinecke's avatar
Martin Reinecke committed
83
    else:
Martin Reinecke's avatar
Martin Reinecke committed
84
        dir = dir * (lin.val.norm()*1e-5/dirder.norm())
Martin Reinecke's avatar
Martin Reinecke committed
85
86
87
88
    # Find a step length that leads to a "reasonable" location
    for i in range(50):
        try:
            loc2 = loc+dir
89
            lin2 = op(Linearization.make_var(loc2, lin.want_metric))
Martin Reinecke's avatar
Martin Reinecke committed
90
91
92
93
94
95
96
97
98
            if np.isfinite(lin2.val.sum()) and abs(lin2.val.sum()) < 1e20:
                break
        except FloatingPointError:
            pass
        dir = dir*0.5
    else:
        raise ValueError("could not find a reasonable initial step")
    return loc2, lin2

Martin Reinecke's avatar
Martin Reinecke committed
99

Martin Reinecke's avatar
Martin Reinecke committed
100
def _check_consistency(op, loc, tol, ntries, do_metric):
Martin Reinecke's avatar
Martin Reinecke committed
101
    for _ in range(ntries):
102
        lin = op(Linearization.make_var(loc, do_metric))
Martin Reinecke's avatar
Martin Reinecke committed
103
        loc2, lin2 = _get_acceptable_location(op, loc, lin)
Martin Reinecke's avatar
Martin Reinecke committed
104
        dir = loc2-loc
Martin Reinecke's avatar
Martin Reinecke committed
105
106
107
108
        locnext = loc2
        dirnorm = dir.norm()
        for i in range(50):
            locmid = loc + 0.5*dir
109
            linmid = op(Linearization.make_var(locmid, do_metric))
Martin Reinecke's avatar
Martin Reinecke committed
110
111
            dirder = linmid.jac(dir)
            numgrad = (lin2.val-lin.val)
Martin Reinecke's avatar
Martin Reinecke committed
112
            xtol = tol * dirder.norm() / np.sqrt(dirder.size)
Martin Reinecke's avatar
Martin Reinecke committed
113
114
            cond = (abs(numgrad-dirder) <= xtol).all()
            if do_metric:
Martin Reinecke's avatar
Martin Reinecke committed
115
116
                dgrad = linmid.metric(dir)
                dgrad2 = (lin2.gradient-lin.gradient)
Martin Reinecke's avatar
Martin Reinecke committed
117
118
                cond = cond and (abs(dgrad-dgrad2) <= xtol).all()
            if cond:
Martin Reinecke's avatar
Martin Reinecke committed
119
120
121
                break
            dir = dir*0.5
            dirnorm *= 0.5
Martin Reinecke's avatar
Martin Reinecke committed
122
            loc2, lin2 = locmid, linmid
Martin Reinecke's avatar
Martin Reinecke committed
123
124
125
        else:
            raise ValueError("gradient and value seem inconsistent")
        loc = locnext
Martin Reinecke's avatar
Martin Reinecke committed
126
127


Martin Reinecke's avatar
Martin Reinecke committed
128
129
130
131
def check_value_gradient_consistency(op, loc, tol=1e-8, ntries=100):
    _check_consistency(op, loc, tol, ntries, False)


Martin Reinecke's avatar
Martin Reinecke committed
132
def check_value_gradient_metric_consistency(op, loc, tol=1e-8, ntries=100):
Martin Reinecke's avatar
Martin Reinecke committed
133
    _check_consistency(op, loc, tol, ntries, True)