Log Message:
Update to 0.4:

0.4

Released on Mar 25th 2013.

    Supports dynamic draw probability.
    Replaces Rating.exposure() with TrueSkill.expose(). Because
    the TrueSkill settings have to adjust a fomula to calculate an
    exposure.
    Deprecates head-to-head shortcut methods in TrueSkill. The
    top-level shortcut functions are still alive.

Version 0.3.1

Released on Mar 6th 2013.

Raises FloatingPointError instead of ValueError (math domain error)
for a problem similar to issue #5 but with more extreme input.

Version 0.3

Released on Mar 5th 2013.

TrueSkill got a new option backend to choose cdf, pdf, ppf
implementation.

When winners have too lower rating than losers, TrueSkill.rate()
will raise FloatingPointError if the backend is None or scipy.
But from this version, you can avoid the problem with mpmath
backend. This was reported at issue #5.

Version 0.2.1

Released on Dec 6th 2012.

Fixes a printing bug on TrueSkill.quality().

Version 0.2

Released on Nov 30th 2012.

    Implements “Partial play”.
    Works well in many Python versions, 2.5, 2.6, 2.7, 3.1, 3.2,
    3.3 and many interpreters, CPython, Jython, PyPy.
    Supports that using dictionaries as a rating_group to choose
    specific playerâs rating simply.
    Adds shorcut functions for 2 players individual match, the most
    usage: rate_1vs1() and quality_1vs1(),
    TrueSkill.transform_ratings() is now called TrueSkill.rate().
    TrueSkill.match_quality() is now called TrueSkill.quality().