README.rst 4.93 KB
Newer Older
Marco Selig's avatar
Marco Selig committed
1
2
NIFTY - Numerical Information Field Theory
==========================================
Marco Selig's avatar
Marco Selig committed
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
30
31
32
33
34
35
36

**NIFTY** project homepage: `<http://www.mpa-garching.mpg.de/ift/nifty/>`_

Summary
-------

Description
...........

**NIFTY**, "\ **N**\umerical **I**\nformation **F**\ield **T**\heor\ **y**\ ",
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 to rapidly
prototype algorithms in 1D, and then apply the developed code in
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.

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.

Marco Selig's avatar
Marco Selig committed
37
*   `Spaces <http://www.mpa-garching.mpg.de/ift/nifty/space.html>`_
Marco Selig's avatar
Marco Selig committed
38

Marco Selig's avatar
Marco Selig committed
39
40
41
42
43
    *   ``point_space`` - unstructured list of points
    *   ``rg_space`` - *n*-dimensional regular Euclidean grid
    *   ``lm_space`` - spherical harmonics
    *   ``gl_space`` - Gauss-Legendre grid on the 2-sphere
    *   ``hp_space`` - `HEALPix <http://sourceforge.net/projects/healpix/>`_
Marco Selig's avatar
Marco Selig committed
44
        grid on the 2-sphere
Marco Selig's avatar
Marco Selig committed
45
    *   ``nested_space`` - arbitrary product of grids
Marco Selig's avatar
Marco Selig committed
46

Marco Selig's avatar
Marco Selig committed
47
*   `Fields <http://www.mpa-garching.mpg.de/ift/nifty/field.html>`_
Marco Selig's avatar
Marco Selig committed
48

Marco Selig's avatar
Marco Selig committed
49
    *   ``field`` - generic class for (discretized) fields::
Marco Selig's avatar
Marco Selig committed
50
51
52
53
54
55

    field.cast_domain   field.hat           field.power        field.smooth
    field.conjugate     field.inverse_hat   field.pseudo_dot   field.tensor_dot
    field.dim           field.norm          field.set_target   field.transform
    field.dot           field.plot          field.set_val      field.weight

Marco Selig's avatar
Marco Selig committed
56
*   `Operators <http://www.mpa-garching.mpg.de/ift/nifty/operator.html>`_
Marco Selig's avatar
Marco Selig committed
57

Marco Selig's avatar
Marco Selig committed
58
59
60
    *   ``diagonal_operator`` - purely diagonal matrices in a specified basis
    *   ``projection_operator`` - projections onto subsets of a specified basis
    *   ``vecvec_operator`` - matrices derived from the outer product of a
Marco Selig's avatar
Marco Selig committed
61
        vector
Marco Selig's avatar
Marco Selig committed
62
    *   ``response_operator`` - exemplary responses that include a convolution,
Marco Selig's avatar
Marco Selig committed
63
        masking and projection
Marco Selig's avatar
Marco Selig committed
64
    *   (and more)
Marco Selig's avatar
Marco Selig committed
65

Marco Selig's avatar
Marco Selig committed
66
* (and more)
Marco Selig's avatar
Marco Selig committed
67
68
69
70
71
72
73
74
75

*Parts of this summary are taken* [1]_.

Installation
------------

Requirements
............

Marco Selig's avatar
Marco Selig committed
76
*   `Python <http://www.python.org/>`_ (v2.7.x)
Marco Selig's avatar
Marco Selig committed
77

Marco Selig's avatar
Marco Selig committed
78
79
80
    *   `NumPy <http://www.numpy.org/>`_ and `SciPy <http://www.scipy.org/>`_
    *   `matplotlib <http://matplotlib.org/>`_
    *   `multiprocessing <http://docs.python.org/2/library/multiprocessing.html>`_
Marco Selig's avatar
Marco Selig committed
81
82
        (standard library)

Marco Selig's avatar
Marco Selig committed
83
*   `GFFT <https://github.com/mrbell/gfft>`_ (v0.1.0) -- Generalized Fast Fourier
Marco Selig's avatar
Marco Selig committed
84
85
    Transformations for Python

Marco Selig's avatar
Marco Selig committed
86
*   `HEALPy <https://github.com/healpy/healpy>`_ (v1.4 without openmp) -- A
Marco Selig's avatar
Marco Selig committed
87
    Python wrapper for `HEALPix <http://sourceforge.net/projects/healpix/>`_
Marco Selig's avatar
Marco Selig committed
88
*   `libsharp-wrapper <https://github.com/mselig/libsharp-wrapper>`_ (v0.1.2
Marco Selig's avatar
Marco Selig committed
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
    without openmp) -- A Python wrapper for the
    `libsharp <http://sourceforge.net/projects/libsharp/>`_ library

Download
........

The latest release is tagged **v0.2.0** and is available as a source package
at `<https://github.com/mselig/nifty/tags>`_. The current version can be
obtained by cloning the repository::

    git clone git://github.com/mselig/nifty.git
    cd nifty

Installation
............

NIFTY is installed using Distutils by running the following command::

    python setup.py install

Alternatively, a private or user specific installation can be done by::

    python setup.py install --user
    python setup.py install --install-lib=/SOMEWHERE

Acknowledgement
---------------

Please, acknowledge the use of NIFTY in your publication(s) by using a phrase
such as the following:

Marco Selig's avatar
Marco Selig committed
120
121
    *"Some of the results in this publication have been derived using the NIFTY
    [M. Selig et al., 2013] package."*
Marco Selig's avatar
Marco Selig committed
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138

References
..........

.. [1] M. Selig et al., "NIFTY -- Numerical Information Field Theory -- a
    versatile Python library for signal inference", submitted to IEEE, 2013;
    `arXiv:XXXX.XXXX <http://www.arxiv.org/abs/XXXX.XXXX>`_

Release Notes
-------------

The NIFTY package is licensed under the
`GPLv3 <http://www.gnu.org/licenses/gpl.html>`_ and is distributed *without any
warranty*.

**NIFTY** project homepage: `<http://www.mpa-garching.mpg.de/ift/nifty/>`_