bayesml 0.2.2

A library for Bayes statistics, Bayes decision theory, and Bayes machine learning

Purpose

BayesML is a library designed for promoting research, education, and application of machine learning based on Bayesian statistics and Bayesian decision theory. Through these activities, BayesML aims to contribute to society.

Characteristics

BayesML has the following characteristics.

• The structure of the library reflects the philosophy of Bayesian statistics and Bayesian decision theory: updating the posterior distribution learned from the data and outputting the optimal estimate based on the Bayes criterion.
• Many of our learning algorithms are much faster than general-purpose Bayesian learning algorithms such as MCMC methods because they effectively use the conjugate property of a probabilistic data generative model and a prior distribution. Moreover, they are suitable for online learning.
• All packages have methods to visualize the probabilistic data generative model, generated data from that model, and the posterior distribution learned from the data in 2~3 dimensional space. Thus, you can effectively understand the characteristics of probabilistic data generative models and algorithms through the generation of synthetic data and learning from them.