README.rst 5.03 KB
Newer Older
Christoph Groth's avatar
Christoph Groth committed
1 2 3
Tinyarray
=========

4 5 6 7 8 9 10
Tinyarrays are similar to NumPy arrays, but optimized for small sizes.
Common operations on very small arrays are to 3-7 times faster than with
NumPy (with NumPy 1.6 it used to be up to 35 times), and 3 times less
memory is used to store them.  Tinyarrays are useful if you need many
small arrays of numbers, and cannot combine them into a few large ones.
(The resulting code is still much slower than C, but it may now be fast
enough.)
Christoph Groth's avatar
Christoph Groth committed
11

12 13 14 15
Unlike Python's built-in tuples, Tinyarrays support mathematical
operations like element-wise addition and matrix multiplication.  Unlike
Numpy arrays, Tinyarrays can be used as dictionary keys because they are
hashable and immutable.
Christoph Groth's avatar
Christoph Groth committed
16

17 18 19
The module's interface is a subset of that of NumPy and thus should be
familiar to many.  Whenever an operation is missing from Tinyarray,
NumPy functions can be used directly with Tinyarrays.
Christoph Groth's avatar
Christoph Groth committed
20 21


22 23
Tinyarray is licensed under the "simplified BSD License".  See
`<LICENSE.rst>`_.
Christoph Groth's avatar
Christoph Groth committed
24

25 26 27 28
Website: https://gitlab.kwant-project.org/kwant/tinyarray

Please report bugs here:
https://gitlab.kwant-project.org/kwant/tinyarray/issues
Christoph Groth's avatar
Christoph Groth committed
29 30


31 32 33 34 35 36 37
Source code
-----------

Source tarballs are available at http://downloads.kwant-project.org/tinyarray/

Clone the Git repository with ::

38
    git clone https://gitlab.kwant-project.org/kwant/tinyarray.git
39 40


41 42
Installation
------------
Christoph Groth's avatar
Christoph Groth committed
43

44
Tinyarray can be built from source with the usual ::
Christoph Groth's avatar
Christoph Groth committed
45

46
    pip install tinyarray
Christoph Groth's avatar
Christoph Groth committed
47

48
One can also download the source tarball (or clone it from git) and use ::
Christoph Groth's avatar
Christoph Groth committed
49

50
    python setup.py install
Christoph Groth's avatar
Christoph Groth committed
51 52

Prepared packages exist for
Christoph Groth's avatar
Christoph Groth committed
53

54
* Windows
Christoph Groth's avatar
Christoph Groth committed
55

56 57
  Use `Christoph Gohlke's installer
  <http://www.lfd.uci.edu/~gohlke/pythonlibs/#tinyarray>`_.
Christoph Groth's avatar
Christoph Groth committed
58

59
* Ubuntu and derivatives ::
60

61 62 63
      sudo apt-add-repository ppa:kwant-project/ppa
      sudo apt-get update
      sudo apt-get install python-tinyarray
64

65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85
* Debian and derivatives

  1. Add the following lines to ``/etc/apt/sources.list``::

         deb http://downloads.kwant-project.org/debian/ stable main
         deb-src http://downloads.kwant-project.org/debian/ stable main

  2. (Optional) Add the OpenPGP key used to sign the repositories by executing::

         sudo apt-key adv --keyserver pgp.mit.edu --recv-key C3F147F5980F3535

  3. Update the package data, and install::

         sudo apt-get update
         sudo apt-get install python-tinyarray

* Mac OS X

  Follow the `instructions for installing "Kwant"
  <http://kwant-project.org/install#mac-os-x>`_ but install
  ``py27-tinyarray`` instead of ``py27-kwant`` etc.
Christoph Groth's avatar
Christoph Groth committed
86 87


88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110
Build configuration
-------------------

If necessary, the compilation and linking of tinyarray can be configured with
a build configuration file.  By default, this file is ``build.conf`` in the
root directory of the tinyarray distribution.  A different path may be
provided using the ``--configfile=PATH`` option.

The configuration file consists of sections, one for each extension module
(currently there is only one: ``tinyarray``), led by a ``[section name]``
header and followed by ``key = value`` lines.

Possible keys are the keyword arguments for ``distutils.core.Extension`` (For
a complete list, see its `documentation
<https://docs.python.org/3/distutils/apiref.html#distutils.core.Extension>`_).
The corresponding values are whitespace-separated lists of strings.

Example ``build.conf`` for compiling Tinyarray with C assertions::

    [tinyarray]
    undef_macros = NDEBUG


111 112
Usage example
-------------
Christoph Groth's avatar
Christoph Groth committed
113

114 115
The following example shows that in simple cases Tinyarray works just as
NumPy. ::
116

117 118
    from math import sin, cos, sqrt
    import tinyarray as ta
119

120 121
    # Make a vector.
    v = ta.array([1.0, 2.0, 3.0])
Christoph Groth's avatar
Christoph Groth committed
122

123 124 125 126 127 128
    # Make a rotation matrix.
    alpha = 0.77
    c, s = cos(alpha), sin(alpha)
    rot_z = ta.array([[c, -s, 0],
                      [s,  c, 0],
                      [0,  0, 1]])
Christoph Groth's avatar
Christoph Groth committed
129

130 131 132 133
    # Rotate the vector, normalize, and print it.
    v = ta.dot(rot_z, v)
    v /= sqrt(ta.dot(v, v))
    print v
Christoph Groth's avatar
Christoph Groth committed
134 135


136 137
Documentation
-------------
Christoph Groth's avatar
Christoph Groth committed
138

139 140 141 142
The module's interface is a basic subset of NumPy and hence should be familiar
to many Python programmers.  All functions are simplified versions of their
NumPy counterparts.  The module's docstring serves as main documentation.  To
see it, run in Python::
Christoph Groth's avatar
Christoph Groth committed
143

144 145
    import tinyarray as ta
    help(ta)
Christoph Groth's avatar
Christoph Groth committed
146

147 148 149 150 151
Or in the system shell::

    pydoc tinyarray


152 153 154 155 156 157 158 159 160 161 162 163 164 165 166
Contributing
------------

Contributions to tinyarray are most welcome.  Patches may be sent by email, or
a merge request may be opened on the Project's website.

Please add tests for any new functionality and make sure that all existing
tests still run.  To run the tests, execute::

    python setup.py test

It is a good idea to enable C assertions as shown above under
`Build configuration`_.


167 168 169
Authors
-------

170
The principal developer of Tinyarray is Christoph Groth (CEA
171 172 173 174 175 176 177
Grenoble).  His contributions are part of his work at `CEA <http://cea.fr/>`_,
the French Commissariat à l'énergie atomique et aux énergies alternatives.

The author can be reached at christoph.groth@cea.fr.

Other people that have contributed to Tinyarray include

178 179
* Michael Wimmer (Leiden University, TU Delft)
* Joseph Weston (CEA Grenoble, TU Delft)