- GRIMS - General R Interface for Markov Sampling.
This is a preliminary release of software for MCMC sampling in R.
- Software for flexible Bayesian modeling
and Markov chain sampling.
This software implements Bayesian models for regression and classification based on neural networks and on Gaussian processes and for Bayesian density estimation and clustering using mixture models and Dirichlet diffusion trees. It also allows you to apply various Markov chain Monte Carlo methods to distributions defined by simple formulas, including simple Bayesian models defined by formulas for the prior and likelihood.

- Software for Low Density Parity Check (LDPC)
codes. Also includes modules for operations on dense and sparse
modulo-2 matrices, and for random number generation.
- Software for Hamiltonian Monte Carlo, and
other MCMC methods, in R.
- An R function for performing univariate slice
sampling.
- Software implementing arithmetic coding for data compression.

- R programs for ensemble MCMC.
- R programs for covariance-adaptive slice sampling.
- Programs for testing Linked Importance Sampling.
- Program to test the short-cut Metropolis method.
- Program to test dragging of fast variables.
- Software for generating data sets from Dirichlet diffusion trees.
- Software for demonstrating prior information transfer.
- Sofware for wake-sleep learning of factor analysis models.
- Scripts and result files for a simplified version of the Bayesian neural network methods I used for the winning entry in the NIPS*2003 feature selections challenge: gzipped tar file (7.6 Megabytes).
- Scripts I used the NIPS*2004 evaluation of predictive uncertainty challenge: tar file.

Back to Radford Neal's home page