.. image:: https://travis-ci.org/timcera/pyslice.svg?branch=master
.. image:: https://coveralls.io/repos/timcera/pyslice/badge.png?branch=master
.. image:: https://img.shields.io/pypi/v/pyslice.svg
:alt: Latest release
.. image:: https://img.shields.io/pypi/dm/pyslice.svg
:alt: PyPI downloads count
.. image:: http://img.shields.io/badge/license-BSD-lightgrey.svg
:alt: pyslice license
Welcome to pyslice - dataset template engine's documentation!
pyslice is a specialized templating system that replaces variables in
a template data set with numbers taken from all combinations of a grouped
series of numbers. It creates a dataset from input template files for each
combination of variables in the series.
The main function of pyslice is to provide utility functions for parametric
modeling. Parametric modeling is a process of varying many inputs to a model.
A drawback to parametric modeling is that there are usually hundreds to
thousands of data sets to prepare and a corresponding number of model runs.
pyslice will create the model data sets and manage the model runs, or place the
model runs in a queue managed by other software. pyslice is also useful in
establishing the sensitivity of a model to changing parameters.
Reference documentation is at :ref:`pyslice_documentation`.
At the command line::
$ pip install pyslice
$ easy_install pyslice
Or, if you have virtualenvwrapper installed::
$ mkvirtualenv pyslice
$ pip install pyslice
Development is managed on bitbucket at
* Reworked the documentation
* Cross compatibility between 2.7 and 3.3 helped out with 'six'.
* Git didn't do what I want with the moves, but
* Moved pyslice.py to pyslice/__init__
* Moved pyslice_lib to pyslice/pyslice_lib
* Ran 'autopep8' on all files - especially useful on PySPG
* Added a README.txt that gives some description.
* Moved to sphinx, restructured text
* Minor improvements to setup.py to allow 'upload_docs' to work.
* Finished transition to git. Deleted CVSROOT and moved everything up
* Demoed all parts of the functionality in the example.
Moved to threads rather than 'os.fork/os.exec' which means that pyslice.py should be able to run on Windows, though not tested. Eliminated a bunch of code required by the 'os.fork/os.exec' that should make pyslice.py easier to maintain.
Uses 'pyslice.ini' instead of 'pyslice.conf'. This allows the use of .ini editors to easily map to the correct format.
Can pull variable values from statistical distributions in Python's 'random' package.
Uses os.path.walk to make considerable faster.
Minor code clean-up
Now depends on Python 2.x or better. Am now using the PySPG library (http://pyspg.sourceforge.net/) in order to develop the variable sets. Because of PySPG now have the capability of doing geometric and list based parameter generation.
Removed Python 2.x dependencies. Variables can now use floating point. Works correctly to just create data sets by setting max_processes = 0 in pyslice.conf. Changes to documentation.
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.