Groups > comp.lang.python.announce > #2124 > unrolled thread

ANN: Lea 2.2.0

Started by	"Pierre Denis" <pie.denis@skynet.be>
First post	2016-05-04 15:57 +0200
Last post	2016-05-04 15:57 +0200
Articles	1 — 1 participant

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

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

  ANN: Lea 2.2.0 "Pierre Denis" <pie.denis@skynet.be> - 2016-05-04 15:57 +0200

#2124 — ANN: Lea 2.2.0

From	"Pierre Denis" <pie.denis@skynet.be>
Date	2016-05-04 15:57 +0200
Subject	ANN: Lea 2.2.0
Message-ID	<mailman.383.1462370685.32212.python-announce-list@python.org>

Lea 2.2.0 is now released!

What is Lea?
------------
Lea is a Python package aiming at working with discrete probability
distributions in an intuitive way. It allows you to model a broad range of
random phenomenons, like dice throwing, coin tossing, gambling, finance,
weather, etc. It offers high-level modeling features for probabilistic
programming and bayesian inference. Lea has several original features: the
storage of probabilities as integer weights, an inference algorithm that
produces *exact* results and a strong emphasis on ease-of-use. Lea is
lightweight, open-source (LGPL) and pure Python, with support of versions 2
and 3). See project page below for installation, tutorials, examples, etc.

What's new in Lea 2.2.0?
------------------------
Compared to latest version (2.1.2), many things have been made to improve
ease-of-use and overall performance. Maybe one of the most notable feature
is that you can now get individual probabilities very easily, as a fraction
or float, thanks to the new 'P' and 'Pf' functions. Here are some examples
that you can type in your Python console:

	>>> P(dice <= 5)
	5/18
	>>> Pf(dice <= 5)
	0.2777777777777778
	>>> P(rain.given(grassWet))
	891/2491
	>>> Pf(rain.given(grassWet))
	0.3576876756322762

Other new features include:
- build joint probability distributions from CSV files or Pandas dataframes
- pmf histograms using matplotlib
- Monte-Carlo sampling estimation
- multi-arguments 'given' method (ANDing of evidences)
- likelihood ratio
- extended 'draw' method: with/without sorting, with/without replacement
- machine learning (experimental)
- built-in functions and distributions for games
- various optimizations

Most of the new features are documented in a new tutorial on Lea's wiki
(http://bitbucket.org/piedenis/lea/wiki/LeaPyTutorial3).

Credits
-------
Thanks to all of you for this large bunch of feedbacks, encouragements and
suggestions! In particular, the present version owes much to Paul Moore, who
made important contributions; among other things, Paul fixed the
installation procedure, set up a test suite using the Tox tool and created
an efficient algorithm for calculating probability distribution resulting
from a drawing process. Thanks Paul for making the package more mature!

Lea project page
----------------
http://bitbucket.org/piedenis/lea

Download Lea (PyPI)
-------------------
http://pypi.python.org/pypi/lea


With the hope that Lea can make your joy less random,

Pierre Denis

[toc] | [standalone]

csiph-web

ANN: Lea 2.2.0

Contents

#2124 — ANN: Lea 2.2.0