~ubuntu-branches/debian/stretch/uncertainties/stretch : revision 1

1

#!/usr/bin/python

2

3

from distutils.core import setup

4

5

setup(name='uncertainties', version='1.4.1',

6

author='Eric O. LEBIGOT (EOL)',

7

author_email='eric.lebigot@normalesup.org',

8

url='http://pypi.python.org/pypi/uncertainties/',

9

10

license='''\

11

This software is released under a dual license. (1) The GNU General \

12

Public License version 2. (2) Any other license, as long as it is \

13

obtained from the original author.''',

14

15

description=('Transparent calculations with uncertainties on the'

16

' quantities involved (aka "error propagation") ;'

17

' calculation of derivatives'),

18

19

long_description=u'''\

20

``uncertainties`` allows calculations such as (2 +/- 0.1)*2 = 4

21

+/- 0.2 to be performed transparently; much more complex mathematical

22

expressions involving numbers with uncertainty can also be evaluated

23

directly.

24

25

**Correlations** between expressions are correctly taken into account.

26

``x-x`` is thus exactly zero, for instance (most implementations

27

found on the web yield a non-zero uncertainty for ``x-x``, which is

28

incorrect).

29

30

Whatever the complexity of the calculation, the number of steps

31

involved, or the correlations between variables, uncertainties

32

produced by this program are what is predicted by `error propagation

33

theory`_.

34

35

36

Basic examples::

37

38

from uncertainties import num_with_uncert

39

from uncertainties.umath import * # sin(), etc.

40

41

# Mathematical operations:

42

x = num_with_uncert((0.20, 0.01)) # x = 0.20+/-0.01

43

x = num_with_uncert("0.20+/-0.01") # Other representation

44

x = num_with_uncert("0.20(1)") # Other representation

45

print x**2 # Square: prints "0.04+/-0.004"

46

print sin(x**2) # Prints "0.0399...+/-0.00399..."

47

48

print x.position_in_sigmas(0.17) # Prints "-3.0": deviation of -3 sigmas

49

50

# Access to the nominal value, and to the uncertainty:

51

square = x**2 # Square

52

print square # Prints "0.04+/-0.004"

53

print square.nominal_value # Prints "0.04"

54

print square.std_dev() # Prints "0.004..."

55

56

print square.derivatives[x] # Partial derivative: prints "0.4" (= 2*0.20)

57

58

print square - x*x # Exactly zero: correlations taken into account

59

60

The Python_ (or IPython_) shell can thus be used as **a powerful

61

calculator** that handles quantities with uncertainties (``print``

62

statements are optional, which is convenient).

63

64

**Almost all mathematical operations** are supported, including most

65

functions from the standard math_ module (sin,...) and functions from the

66

third-party NumPy_ module (fast operations on arrays and matrices).

67

Comparison operators (``>``, ``==``, etc.) are supported too. There is

68

no restriction on the complexity of mathematical expressions, or on the

69

number of variables involved (x-sin(x)+y**2-tan(y*x) can for example be

70

calculated, whether x and y are quantities with uncertainties or not).

71

72

Another possible use of this module is the calculation of **partial

73

derivatives** of mathematical functions (they are used by `error

74

propagation theory`_, and are thus automatically calculated by this

75

module).

76

77

Additional examples and information can be obtained with ``pydoc

78

uncertainties`` and ``pydoc uncertainties.umath`` after installation.

79

80

*Installation*: ``sudo easy_install uncertainties`` might be

81

sufficient, depending on your installation (this does not require any

82

manual download, but requires setuptools_). For additional

83

installation methods, download the source archive, and see the

84

``README.txt`` file that it contains.

85

86

*User feedback*:

87

88

- "*A gift of the gods for the work I\'m doing*" (e-mail)

89

- "*Holy f\*\*\* this would have saved me so much f\*\*\*ing time last\

90

semester*" (reddit_)

91

- "*Very useful*" (reddit_)

92

93

Please send feature requests, bug reports, or feedback to

94

`Eric O. LEBIGOT (EOL)`_.

95

96

*Version history* (main changes only):

97

98

- 1.4: added three utilities for manipulating NumPy arrays of numbers with\

99

uncertainty (see the main package documentation).

100

- 1.3: numbers with uncertainty are now constructed with \

101

``num_with_uncert()``, which replaces ``NumberWithUncert()``. This \

102

simplifies the class hierarchy by removing the ``NumberWithUncert`` class.

103

- 1.2.5: numbers with uncertainty can now be entered as \

104

``NumberWithUncert("1.23+/-0.45")`` too.

105

- 1.2.3: ``log(x, base)`` is now supported by ``umath.log()``, in addition \

106

to ``log(x)``.

107

- 1.2.2: values with uncertainties are now output like 3+/-1, in order \

108

to avoid confusing 3+-1 with 3+(-1).

109

- 1.2: a new function, ``wrap()``, is exposed, which allows non-Python \

110

functions (e.g. Fortran or C used through a module such as SciPy) to \

111

handle numbers with uncertainty.

112

- 1.1: mathematical functions (such as cosine, etc.) are in a new \

113

uncertainties.umath module; \

114

they do not override functions from the math module anymore.

115

- 1.0.12: main class (``Number_with_uncert``) renamed ``NumberWithUncert`` \

116

so as to follow `PEP 8`_.

117

- 1.0.11: ``origin_value`` renamed more appropriately as \

118

``nominal_value``.

119

- 1.0.9: ``correlations()`` renamed more appropriately as \

120

``covariance_matrix()``.

121

122

.. _Python: http://docs.python.org/tutorial/interpreter.html

123

.. _IPython: http://ipython.scipy.org/

124

.. _NumPy: http://numpy.scipy.org/

125

.. _math: http://docs.python.org/library/math.html

126

.. _PEP 8: http://www.python.org/dev/peps/pep-0008/

127

.. _error propagation theory: http://en.wikipedia.org/wiki/Propagation\

128

_of_uncertainty

129

.. _setuptools: http://pypi.python.org/pypi/setuptools

130

.. _Eric O. LEBIGOT (EOL): mailto:eric.lebigot@normalesup.org

131

.. _reddit: http://www.reddit.com/r/Python/comments/am84v/\

132

now_you_can_do_calculations_with_uncertainties_5/

133

''',

134

135

keywords=['error propagation', 'uncertainties',

136

'uncertainty calculations',

137

'standard deviation',

138

'derivatives', 'partial derivatives', 'differentiation'],

139

140

classifiers=[

141

'Development Status :: 5 - Production/Stable',

142

'Intended Audience :: Developers',

143

'Intended Audience :: Education',

144

'Intended Audience :: Other Audience',

145

'Intended Audience :: Science/Research',

146

'Operating System :: OS Independent',

147

'Programming Language :: Python',

148

'Programming Language :: Python :: 2.5',

149

'Programming Language :: Python :: 2.6',

150

'Topic :: Scientific/Engineering',

151

'Topic :: Scientific/Engineering :: Mathematics',

152

'Topic :: Scientific/Engineering :: Physics',

153

'Topic :: Software Development',

154

'Topic :: Software Development :: Libraries',

155

'Topic :: Software Development :: Libraries :: Python Modules',

156

'Topic :: Utilities'

157

],

158

159

# Files are defined in MANIFEST.in

160

packages=['uncertainties']

161

# py_modules=['uncertainties', 'uncertainties.umath']

162

)

163

164