smooth_linear_amplitude.py 8.39 KB
Newer Older
Martin Reinecke's avatar
Martin Reinecke committed
1 2 3 4 5 6 7 8 9 10 11 12 13
# 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/>.
#
14
# Copyright(C) 2013-2019 Max-Planck-Society
Martin Reinecke's avatar
Martin Reinecke committed
15
#
16
# NIFTy is being developed at the Max-Planck-Institut fuer Astrophysik.
Philipp Arras's avatar
Philipp Arras committed
17

18
import numpy as np
Philipp Arras's avatar
Philipp Arras committed
19

Philipp Arras's avatar
Docs  
Philipp Arras committed
20
from ..domain_tuple import DomainTuple
Martin Reinecke's avatar
Martin Reinecke committed
21
from ..domains.power_space import PowerSpace
22
from ..field import Field
Philipp Arras's avatar
Philipp Arras committed
23
from ..operators.adder import Adder
Philipp Arras's avatar
Philipp Arras committed
24 25 26 27
from ..operators.exp_transform import ExpTransform
from ..operators.qht_operator import QHTOperator
from ..operators.slope_operator import SlopeOperator
from ..operators.symmetrizing_operator import SymmetrizingOperator
28
from ..sugar import makeOp
29

30 31 32 33 34 35
from ..library.inverse_gamma_operator import InverseGammaOperator
from ..sugar import from_global_data
from ..operators.diagonal_operator import DiagonalOperator
from ..operators.simple_linear_operators import FieldAdapter
from ..operators.value_inserter import ValueInserter

36

37
def _ceps_kernel(k, a, k0):
Philipp Arras's avatar
Philipp Arras committed
38
    return (a/(1 + np.sum((k.T/k0)**2, axis=-1).T))**2
39 40


Philipp Arras's avatar
Docs  
Philipp Arras committed
41
def CepstrumOperator(target, a, k0):
42
    """Turns a white Gaussian random field into a smooth field on a LogRGSpace.
43

Philipp Arras's avatar
Docs  
Philipp Arras committed
44 45 46 47 48 49 50 51
    Composed out of three operators:

        sym @ qht @ diag(sqrt_ceps),

    where sym is a :class:`SymmetrizingOperator`, qht is a :class:`QHTOperator`
    and ceps is the so-called cepstrum:

    .. math::
Martin Reinecke's avatar
Martin Reinecke committed
52
        \\mathrm{sqrt\\_ceps}(k) = \\frac{a}{1+(k/k0)^2}
Philipp Arras's avatar
Docs  
Philipp Arras committed
53 54 55 56 57 58 59

    These operators are combined in this fashion in order to generate:

    - A field which is smooth, i.e. second derivatives are punished (note
      that the sqrt-cepstrum is essentially proportional to 1/k**2).

    - A field which is symmetric around the pixel in the middle of the space.
Philipp Arras's avatar
Docs  
Philipp Arras committed
60 61
      This is result of the :class:`SymmetrizingOperator` and needed in order
      to decouple the degrees of freedom at the beginning and the end of the
Philipp Arras's avatar
Docs  
Philipp Arras committed
62 63 64
      amplitude whenever :class:`CepstrumOperator` is used as in
      :class:`SLAmplitude`.

Philipp Arras's avatar
Docs  
Philipp Arras committed
65 66
    The prior on the zero mode (or zero subspaces for more than one dimensions)
    is the integral of the prior over all other modes along the corresponding
67
    axis.
Philipp Arras's avatar
Docs  
Philipp Arras committed
68 69 70 71

    Parameters
    ----------
    target : LogRGSpace
72
        Target domain of the operator, needs to be non-harmonic.
Philipp Arras's avatar
Docs  
Philipp Arras committed
73
    a : float
74 75 76 77
        Cutoff of smoothness prior (positive only). Controls the
        regularization of the inverse laplace operator to be finite at zero.
        Larger values for the cutoff results in a weaker constraining prior.
    k0 : float, list of float
78
        Strength of smoothness prior in quefrency space (positive only) along
79 80
        each axis. If float then the strength is the same along each axis.
        Larger values result in a weaker constraining prior.
81
    """
82
    a = float(a)
Philipp Arras's avatar
Docs  
Philipp Arras committed
83
    target = DomainTuple.make(target)
84
    if a <= 0:
Philipp Arras's avatar
Docs  
Philipp Arras committed
85
        raise ValueError
86
    if len(target) > 1 or target[0].harmonic:
Philipp Arras's avatar
Docs  
Philipp Arras committed
87
        raise TypeError
Philipp Arras's avatar
Docs  
Philipp Arras committed
88 89 90 91 92
    if isinstance(k0, (float, int)):
        k0 = np.array([k0]*len(target.shape))
    else:
        k0 = np.array(k0)
    if len(k0) != len(target.shape):
93 94 95
        raise ValueError
    if np.any(np.array(k0) <= 0):
        raise ValueError
Philipp Arras's avatar
Docs  
Philipp Arras committed
96 97 98 99 100 101 102 103 104

    qht = QHTOperator(target)
    dom = qht.domain[0]
    sym = SymmetrizingOperator(target)

    # Compute cepstrum field
    dim = len(dom.shape)
    shape = dom.shape
    q_array = dom.get_k_array()
105
    # Fill all non-zero modes
Philipp Arras's avatar
Philipp Arras committed
106 107
    no_zero_modes = (slice(1, None),)*dim
    ks = q_array[(slice(None),) + no_zero_modes]
108
    cepstrum_field = np.zeros(shape)
Philipp Arras's avatar
Docs  
Philipp Arras committed
109
    cepstrum_field[no_zero_modes] = _ceps_kernel(ks, a, k0)
110
    # Fill zero-mode subspaces
111
    for i in range(dim):
Philipp Arras's avatar
Philipp Arras committed
112 113 114
        fst_dims = (slice(None),)*i
        sl = fst_dims + (slice(1, None),)
        sl2 = fst_dims + (0,)
115
        cepstrum_field[sl2] = np.sum(cepstrum_field[sl], axis=i)
Philipp Arras's avatar
Docs  
Philipp Arras committed
116
    cepstrum = Field.from_global_data(dom, cepstrum_field)
Philipp Arras's avatar
Philipp Arras committed
117

118 119 120
    return sym @ qht @ makeOp(cepstrum.sqrt())


121
def SLAmplitude(*, target, n_pix, a, k0, sm, sv, im, iv, za=None, zq=None,
122
                keys=['tau', 'phi', 'zero_mode']):
123 124
    '''Operator for parametrizing smooth amplitudes (square roots of power
    spectra).
125 126 127

    The general guideline for setting up generative models in IFT is to
    transform the problem into the eigenbase of the prior and formulate the
128 129
    generative model in this base. This is done here for the case of an
    amplitude which is smooth and has a linear component (both on
130 131 132
    double-logarithmic scale).

    This function assembles an :class:`Operator` which maps two a-priori white
133
    Gaussian random fields to a smooth amplitude which is composed out of
134 135 136 137 138 139
    a linear and a smooth component.

    On double-logarithmic scale, i.e. both x and y-axis on logarithmic scale,
    the output of the generated operator is:

        AmplitudeOperator = 0.5*(smooth_component + linear_component)
Philipp Arras's avatar
Philipp Arras committed
140

141
    This is then exponentiated and exponentially binned (in this order).
142 143 144 145 146 147

    The prior on the linear component is parametrized by four real numbers,
    being expected value and prior variance on the slope and the y-intercept
    of the linear function.

    The prior on the smooth component is parametrized by two real numbers: the
Martin Reinecke's avatar
Martin Reinecke committed
148 149
    strength and the cutoff of the smoothness prior
    (see :class:`CepstrumOperator`).
Martin Reinecke's avatar
Martin Reinecke committed
150 151 152

    Parameters
    ----------
153 154 155 156 157 158 159
    n_pix : int
        Number of pixels of the space in which the .
    target : PowerSpace
        Target of the Operator.
    a : float
        Strength of smoothness prior (see :class:`CepstrumOperator`).
    k0 : float
Philipp Arras's avatar
Docs  
Philipp Arras committed
160 161
        Cutoff of smothness prior in quefrency space (see
        :class:`CepstrumOperator`).
Philipp Arras's avatar
Philipp Arras committed
162
    sm : float
Philipp Arras's avatar
Docs  
Philipp Arras committed
163
        Expected exponent of power law.
Philipp Arras's avatar
Philipp Arras committed
164
    sv : float
165
        Prior standard deviation of exponent of power law.
Philipp Arras's avatar
Philipp Arras committed
166
    im : float
Philipp Arras's avatar
Docs  
Philipp Arras committed
167 168
        Expected y-intercept of power law. This is the value at t_0 of the
        LogRGSpace (see :class:`ExpTransform`).
Philipp Arras's avatar
Philipp Arras committed
169
    iv : float
170
        Prior standard deviation of y-intercept of power law.
171
    za : float, optional
172 173
        The alpha-parameter of the inverse-gamma distribution.
        Setting the a seperate prior on the zeroGmode of the amplitude model.
174
    zq : float, optional
175
        The q-parameter of the inverse-gamma distribution.
176 177 178 179 180 181 182

    Returns
    -------
    Operator
        Operator which is defined on the space of white excitations fields and
        which returns on its target a power spectrum which consists out of a
        smooth and a linear part.
Martin Reinecke's avatar
Martin Reinecke committed
183
    '''
184 185 186 187 188
    if not (isinstance(n_pix, int) and isinstance(target, PowerSpace)):
        raise TypeError

    a, k0 = float(a), float(k0)
    sm, sv, im, iv = float(sm), float(sv), float(im), float(iv)
189 190
    if sv <= 0 or iv <= 0:
        raise ValueError
191

192
    if za is not None and zq is not None:
193 194 195 196
        separate_zero_mode_prior = True
        za, zq = float(za), float(zq)
    else:
        separate_zero_mode_prior = False
197
        if za is not None or zq is not None:
198 199
            raise ValueError("za and zq have to be given together")

200 201 202
    et = ExpTransform(target, n_pix)
    dom = et.domain[0]

203
    # Smooth component
204 205
    dct = {'a': a, 'k0': k0}
    smooth = CepstrumOperator(dom, **dct).ducktape(keys[0])
Martin Reinecke's avatar
Martin Reinecke committed
206

207
    # Linear component
208 209 210 211 212
    sl = SlopeOperator(dom)
    mean = np.array([sm, im + sm*dom.t_0[0]])
    sig = np.array([sv, iv])
    mean = Field.from_global_data(sl.domain, mean)
    sig = Field.from_global_data(sl.domain, sig)
Philipp Arras's avatar
Philipp Arras committed
213
    linear = sl @ Adder(mean) @ makeOp(sig).ducktape(keys[1])
214 215 216

    # Combine linear and smooth component
    loglog_ampl = 0.5*(smooth + linear)
Philipp Arras's avatar
Changes  
Philipp Arras committed
217

218 219 220 221 222 223 224 225 226 227 228 229 230 231 232
    if not separate_zero_mode_prior:
        # Go from loglog-space to linear-linear-space
        return et @ loglog_ampl.exp()
    else:
        zero_mode = ValueInserter(et.target, (0,)*len(et.target.shape))

        mask = np.ones(et.target.shape)
        mask[(0,)*len(et.target.shape)] = 0.
        mask = from_global_data(et.target, mask)
        mask = DiagonalOperator(mask)
        zero_mode = zero_mode @ InverseGammaOperator(
            zero_mode.domain, za, zq) @ FieldAdapter(
                zero_mode.domain, keys[2])

        return mask @ (et @ loglog_ampl.exp()) + zero_mode