dysii Dynamic Systems Library

This C++ library provides a collection of classes useful for probabilistic machine learning within dynamic systems. It features the following:

Filtering and smoothing

  • Kalman filter and smoother
  • Rauch-Tung-Striebel (RTS) smoother
  • Unscented Kalman filter and smoother
  • Particle filter and forward-backward smoother
  • Multiple resampling strategies for particle filters, including stratified and auxiliary resampling

Probability distributions

  • Gaussian and Gaussian mixture distributions
  • Dirac and Dirac mixture (weighted sample set) distributions
  • Density tree and kernel density (kd) trees

Differential equations

  • Stochastic and deterministic Runge-Kutta numerical solvers with adaptive time step

Parallelisation

  • Parallel particle filter and smoother
  • Distributed storage of mixtures

Serialization

  • Serialization of probability distributions for convenient data management

Performance

  • Use of BLAS and LAPACK
  • Template meta-programming
  • Code profiling
  • Compiler optimisation

Features are added to the library as needed, so that it does lack some completeness. What is included, however, is well documented and tested, and may be considered sound for research purposes.

The library has been optimised for performance, while maintaining a modularity and generality that makes it suitable for a wide range of applications.


Related posts

dysii 1.3.0 released

Wednesday, March 5th, 2008

Version 1.3.0 of the dysii Dynamic Systems Library is now available. The new release adds a stochastic Runge-Kutta method for stochastic differential systems, and preliminary density and kernel density (KD) tree implementations.
See the updated documentation and VERSION.txt file in the new distribution for more information.
Both a detailed tutorial and installation guide are on the way […]

dysii 1.2.0 Released

Sunday, December 2nd, 2007

Version 1.2.0 of the dysii Dynamic Systems Library is now available. The new release adds the Auxiliary Particle Filter, a generalised framework for particle filter resampling strategies, and makes a number of optimisations and bug fixes.
Specific changes include:

Added auxiliary particle filter.
Added generalised resampling strategy framework.
Fixed diagonal covariance detection for optimised Gaussian density calculations.
Fixed several serialization […]

Continuous Time Particle Filtering for fMRI

Thursday, November 29th, 2007

I’ve added a draft version of the paper Continuous Time Particle Filtering for fMRI to the research page, to be presented as a poster at NIPS 2007. Note that this version of the paper is not available in the conference pre-proceedings. The work makes use of both the dysii Dynamic Systems Library for distributed/parallel particle […]

Related feeds