Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.lang.python.announce > #3197 > unrolled thread

tsplib95 v0.4.0

Started by"Robert Grant" <rhgrant10@gmail.com>
First post2019-09-23 00:25 +0000
Last post2019-09-23 00:25 +0000
Articles 1 — 1 participant

Back to article view | Back to comp.lang.python.announce


Contents

  tsplib95 v0.4.0 "Robert Grant" <rhgrant10@gmail.com> - 2019-09-23 00:25 +0000

#3197 — tsplib95 v0.4.0

From"Robert Grant" <rhgrant10@gmail.com>
Date2019-09-23 00:25 +0000
Subjecttsplib95 v0.4.0
Message-ID<156919833221.7290.15347002846613920338@mail.python.org>
I'm pleased to announce the release of tsplib95 v0.4.0!

What's New?
-----------

* parser now raises ``ParserError`` instead of ``Exception``
* problem fields are now documented
* added explicit support for python 3.7
* corrected a parameter name in ``distances.geographical``

What is tsplib95?
-----------------

tsplib95 is a library for reading TSPLIB95 problems and solutions.
It supports the entire file format specification and automatically
converts problems into ``networkx.Graph`` instances.

License: Apache Software License 2.0

* Read the **docs**: https://tsplib95.readthedocs.io/
* Install from **PyPI**: https://pypi.org/project/tsplib95/0.4.0/
* Contribute **source**: https://github.com/rhgrant10/tsplib95

Example
-------

.. code-block:: python

    >>> import tsplib95
    >>> import networkx as nx

    >>> # read problem files
    >>> problem = tsplib95.load_problem('path/to/att48.tsp')
    >>> problem.name
    >>> problem.comment
    '48 capitals of the US (Padberg/Rinaldi)'
    >>> problem.dimension
    48
    >>> problem.type
    'TSP'
    
    >>> # auto-generated weight function for any two nodes
    >>> problem.wfunc(2, 6)
    6977

    >>> # convert to networkx
    >>> G = problem.get_graph()
    >>> if problem.is_symmteric():
    ...     assert issubclass(G, nx.Graph)
    ... else:
    ...     assert issubclass(G, nx.DiGraph)
    >>> if problem.is_depictable():
    ...     nx.draw(G)
    ... 


Enjoy!

Robert Grant

[toc] | [standalone]


Back to top | Article view | comp.lang.python.announce


csiph-web