stats-arrays

The stats_arrays package provides a standard NumPy array interface for defining uncertain parameters used in models, and classes for Monte Carlo sampling. It also plays well with others.

Motivation

Want a consistent interface to SciPy and NumPy statistical function
Want to be able to quickly load and save many parameter uncertainty distribution definitions in a portable format
Want to manipulate and switch parameter uncertainty distributions and variables
Want simple Monte Carlo random number generators that return a vector of parameter values to be fed into uncertainty or sensitivity analysis
Want something simple, extensible, documented and tested

The stats_arrays package was originally developed for the Brightway2 life cycle assessment framework, but can be applied to any stochastic model.

Example

>>> from stats_arrays import *
>>> my_variables = UncertaintyBase.from_dicts(
...     {'loc': 2, 'scale': 0.5, 'uncertainty_type': NormalUncertainty.id},
...     {'loc': 1.5, 'minimum': 0, 'maximum': 10, 'uncertainty_type': TriangularUncertainty.id}
... )
>>> my_variables
array([(2.0, 0.5, nan, nan, nan, False, 3),
       (1.5, nan, nan, 0.0, 10.0, False, 5)],
    dtype=[('loc', '<f8'), ('scale', '<f8'), ('shape', '<f8'),
           ('minimum', '<f8'), ('maximum', '<f8'), ('negative', '?'),
           ('uncertainty_type', 'u1')])
>>> my_rng = MCRandomNumberGenerator(my_variables)
>>> my_rng.next()
array([ 2.74414022,  3.54748507])
>>> # can also be used as an interator
>>> zip(my_rng, xrange(10))
[(array([ 2.96893108,  2.90654471]), 0),
 (array([ 2.31190619,  1.49471845]), 1),
 (array([ 3.02026168,  3.33696367]), 2),
 (array([ 2.04775418,  3.68356226]), 3),
 (array([ 2.61976694,  7.0149952 ]), 4),
 (array([ 1.79914025,  6.55264372]), 5),
 (array([ 2.2389968 ,  1.11165296]), 6),
 (array([ 1.69236527,  3.24463981]), 7),
 (array([ 1.77750176,  1.90119991]), 8),
 (array([ 2.32664152,  0.84490754]), 9)]

Source code: https://bitbucket.org/cmutel/stats_arrays
Online documentation: https://stats_arrays.readthedocs.org/en/latest/

Metadata

Release files for stats-arrays 0.2.2

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for stats-arrays 0.2.2
File	Size	Uploaded
stats_arrays-0.2.2.tar.gz	27.6 kB	Sep 6, 2013	Details

Release files / stats_arrays-0.2.2.tar.gz

Download URL	stats_arrays-0.2.2.tar.gz
Size	27.6 kB
Tags	Source
SHA-256 checksum How to use checksums	`58d295f5ec48a95343f53c4127ef7c1cb7caa6553dd4fb94108111c9df26ec2e`
BLAKE2b-256 checksum How to use checksums	`421ef9e302f9943a25d65584db46eebf98e877f4e28f8ee60e7967ac811415c5`
Upload date	Sep 6, 2013
Uploaded using Trusted Publishing? What is trusted publishing?	No