ChangeLog 6.17 KB
Newer Older
Martin Reinecke's avatar
Martin Reinecke committed
1
2
Changes since NIFTy 5:

Martin Reinecke's avatar
Martin Reinecke committed
3
4
5
6
7
8
9
FFT convention adjusted
=======================

When going to harmonic space, NIFTy's FFT operator now uses a minus sign in the
exponent (and, consequently, a plus sign on the adjoint transform). This
convention is consistent with almost all other numerical FFT libraries.

Martin Reinecke's avatar
Martin Reinecke committed
10
11
12
Interface change in EndomorphicOperator.draw_sample()
=====================================================

Philipp Arras's avatar
Philipp Arras committed
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Both complex-valued and real-valued Gaussian probability distributions have
hermitian and positive endomorphisms as covariance. Just by looking at an
endomorphic operator itself it is not clear whether it is viewed as covariance
for real or complex Gaussians when a sample of the respective distribution shall
be drawn. Therefore, we introduce the method `draw_sample_with_dtype()` which
needs to be given the data type of the probability distribution. This function
is implemented for all operators which actually draw random numbers
(`DiagonalOperator` and `ScalingOperator`). The class `SamplingDtypeSetter` acts
as a wrapper for these kind of operators in order to fix the data type of the
distribution. Samples from these operators can be drawn with `.draw_sample()`.
In order to dive into those subtleties I suggest running the following code and
playing around with the dtypes.

```
import nifty6 as ift
import numpy as np

dom = ift.UnstructuredDomain(5)
dtype = [np.float64, np.complex128][1]
invcov = ift.ScalingOperator(dom, 3)
e = ift.GaussianEnergy(mean=ift.from_random('normal', dom, dtype=dtype),
                       inverse_covariance=invcov)
pos = ift.from_random('normal', dom, dtype=np.complex128)
lin = e(ift.Linearization.make_var(pos, want_metric=True))
met = lin.metric
print(met)
print(met.draw_sample())
```
Martin Reinecke's avatar
Martin Reinecke committed
41

Philipp Arras's avatar
Philipp Arras committed
42
43
44
45
46
47
MPI parallelisation over samples in MetricGaussianKL
====================================================

The classes `MetricGaussianKL` and `MetricGaussianKL_MPI` have been unified
into one `MetricGaussianKL` class which has MPI support built in.

Martin Reinecke's avatar
Martin Reinecke committed
48
49
50
51
52
53
54
55
56
New approach for random number generation
=========================================

The code now uses `numpy`'s new `SeedSequence` and `Generator` classes for the
production of random numbers (introduced in numpy 1.17. This greatly simplifies
the generation of reproducible random numbers in the presence of MPI parallelism
and leads to cleaner code overall. Please see the documentation of
`nifty6.random` for details.

Philipp Arras's avatar
Philipp Arras committed
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
Interface Change for non-linear Operators
=========================================

The method `Operator.apply()` takes a `Linearization` or a `Field` or a
`MultiField` as input. This has not changed. However, now each non-linear
operator assumes that the input `Linearization` comes with an identity operator
as jacobian. Also it is assumed that the `apply()` method returns a
`Linearization` with the jacobian of the operator itself. The user is not in
charge anymore of stacking together the jacobians of operator chains. Something
like `x.jac` should not appear in any self-written `apply()` methods. The method
`Operator._check_input` tests if this condition is met. The same goes for the
metric. There is no need anymore to call `SandwichOperator` in an `apply()`
method when implementing new energies. This change should not lead to unexpected
behaviour since both `Operator._check_input()` and
`extra.check_jacobian_consistency()` tests for the new conditions to be
fulfilled.

Martin Reinecke's avatar
Martin Reinecke committed
74
75
76
77
78
79
80
81
82
83
84
85
Special functions for complete Field reduction operations
=========================================================

So far, reduction operations called on Fields (like `vdot`, `sum`, `integrate`,
`mean`, `var`, `std`, `prod` etc.) returned a scalar when the reduction was
carried out over all domains, and otherwise a `Field`.
Having the data type of the returned value depend on input parameters is
extremely confusing, so all of these reduction operations now always return a
Field. We also introduced another set of reduction operations which always
operate over all subdomains and therefore don't take a `spaces` argument; they
are named `s_vdot`, `s_sum` etc. and always return a scalar.

86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
Updates regarding correlated fields
===================================

The most commonly used model for homogeneous and isotropic correlated fields in
nifty5 has been `SLAmplitude` combined with `CorrelatedField`. This model
exhibits unintuitive couplings between its parameters and as been replaced
by `CorrelatedFieldMaker` in NIFTy 6. This model aims to conceptionally provide
the same functionality. However, internally it works quite differently. Therefore,
specific classes for `SLAmplitude` like `LogRGSpace`, `QHTOperator`, `ExpTransform`,
`SlopeOperator`, `SymmetrizingOperator`, `CepstrumOperator`, `CorrelatedField`
and `MfCorrelatedField` are not needed anymore and have been removed. In general,
`CorrelatedFieldMaker` feels to be better conditioned leading to faster convergence
but it is hard to make explicit tests since the two approaches cannot be mapped
onto each other exactly. We experienced that preconditioning in the `MetricGaussianKL`
via `napprox` breaks the inference scheme with the new model so `napprox` may not
be used here.
Martin Reinecke's avatar
merge    
Martin Reinecke committed
102

Lukas Platz's avatar
Lukas Platz committed
103
104
Removal of the standard MPI parallelization scheme:
===================================================
Martin Reinecke's avatar
Martin Reinecke committed
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121

When several MPI tasks are present, NIFTy5 distributes every Field over these
tasks by splitting it along the first axis. This approach to parallelism is not
very efficient, and it has not been used by anyone for several years, so we
decided to remove it, which led to many simplifications within NIFTy.
User-visible changes:
- the methods `to_global_data`, `from_global_data`, `from_local_data` and
  the property `local_data` have been removed from `Field` and `MultiField`.
  Instead there are now the property `val` (returning a read-only numpy.ndarray
  for `Field` and a dictionary of read-only numpy.ndarrays for `MultiField`) and
  the method `val_rw()` (returning the same structures with writable copies of
  the arrays). Fields and MultiFields can be created from such structures using
  the static method `from_raw`
- the functions `from_global_data` and `from_local_data` in `sugar` have been
  replaced by a single function called `makeField`
- the property `local_shape` has been removed from `Domain` (and subclasses)
  and `DomainTuple`.