dysii is a C++ library for distributed probabilistic inference and learning in large-scale dynamical systems. It provides methods such as the Kalman, unscented Kalman, and particle filters and smoothers, as well as useful classes such as common probability distributions and stochastic processes.
|Tags||Scientific/Engineering Artificial Intelligence Software Development Libraries|
Release Notes: This release adds kernel density estimators with distributed kd tree partitioning and dual-tree evaluations, an improved stochastic Runge-Kutta and new Euler-Maruyama integrator for stochastic differential equations, the kernel forward-backward and two-filter smoothers (from the author's PhD work), performance enhancements, and an installation guide.
Release Notes: This version adds a stochastic Runge-Kutta method for stochastic differential equations, as well as density and kernel density (KD) trees for representing probability densities.
Release Notes: An auxiliary particle filter was added and the resampling strategy framework was generalized. Diagonal covariance detection for optimized Gaussian density calculations was fixed. Several serialization bugs and a Wiener process variance bug were fixed.
Release Notes: Overhauled parallel implementations. The particle smoother has been improved with further parallelisation. Distributed storage of mixtures has been added, as well as Gaussian mixture distributions and serialization of probability distributions. A Wiener process variance bug has been fixed.
No changes have been submitted for this release.