README.rst 5.04 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

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

Marco Selig's avatar
Marco Selig committed
6 7
**!!! THIS PAGE IS CURRENTLY UNDER CONSTRUCTION !!!**

Marco Selig's avatar
Marco Selig committed
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 37 38
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
39
*   `Spaces <http://www.mpa-garching.mpg.de/ift/nifty/space.html>`_
Marco Selig's avatar
Marco Selig committed
40

Marco Selig's avatar
Marco Selig committed
41 42 43 44 45
    *   ``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
46
        grid on the 2-sphere
Marco Selig's avatar
Marco Selig committed
47
    *   ``nested_space`` - arbitrary product of grids
Marco Selig's avatar
Marco Selig committed
48

Marco Selig's avatar
Marco Selig committed
49
*   `Fields <http://www.mpa-garching.mpg.de/ift/nifty/field.html>`_
Marco Selig's avatar
Marco Selig committed
50

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

::
Marco Selig's avatar
Marco Selig committed
54 55 56 57 58 59

    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
60
*   `Operators <http://www.mpa-garching.mpg.de/ift/nifty/operator.html>`_
Marco Selig's avatar
Marco Selig committed
61

Marco Selig's avatar
Marco Selig committed
62 63 64
    *   ``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
65
        vector
Marco Selig's avatar
Marco Selig committed
66
    *   ``response_operator`` - exemplary responses that include a convolution,
Marco Selig's avatar
Marco Selig committed
67
        masking and projection
Marco Selig's avatar
Marco Selig committed
68
    *   (and more)
Marco Selig's avatar
Marco Selig committed
69

Marco Selig's avatar
Marco Selig committed
70
* (and more)
Marco Selig's avatar
Marco Selig committed
71

Marco Selig's avatar
Marco Selig committed
72 73
*Parts of this summary are taken* [1]_ *without marking them explicitly as
quotations.*
Marco Selig's avatar
Marco Selig committed
74 75 76 77 78 79 80

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

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

Marco Selig's avatar
Marco Selig committed
81
*   `Python <http://www.python.org/>`_ (v2.7.x)
Marco Selig's avatar
Marco Selig committed
82

Marco Selig's avatar
Marco Selig committed
83 84 85
    *   `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
86 87
        (standard library)

Marco Selig's avatar
Marco Selig committed
88 89
*   `GFFT <https://github.com/mrbell/gfft>`_ (v0.1.0) -- Generalized Fast
    Fourier Transformations for Python
Marco Selig's avatar
Marco Selig committed
90

Marco Selig's avatar
Marco Selig committed
91
*   `HEALPy <https://github.com/healpy/healpy>`_ (v1.4 without openmp) -- A
Marco Selig's avatar
Marco Selig committed
92
    Python wrapper for `HEALPix <http://sourceforge.net/projects/healpix/>`_
Marco Selig's avatar
Marco Selig committed
93
*   `libsharp-wrapper <https://github.com/mselig/libsharp-wrapper>`_ (v0.1.2
Marco Selig's avatar
Marco Selig committed
94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124
    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
125 126
    *"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
127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143

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/>`_