Changed ddirich to calculate on log scale rather than logging after. Changed starting tuning values for r's. Fixed lambda_j_prop calculation (didn't work if lambda j not calculated first)

Fixed problem with initial starting values for q being in the wrong order. Changed update lists to a single list to remove mcmc func closures. Added names to updaters so they can be identified for the acceptance rates. Added some more help file notes. NOT WORKING CORRECTLY for sim data.

Merging my updates to branch 'interface' of fhm-chicas-code.lancs.ac.uk:jewell/sourceR-G into interface.

Added AdaptiveSingleSiteMRW updater.  Corrected a problem in AdaptiveLogDirMRW where individual components were not being able to adapt independently.

Small change to PoissonNode to deal with cases when the length of the mean value is different to the offset.  Usual R conventions apply here.

Implemented AdaptiveLogDirMRW2 and DirichletNode2 which reparameterises the Dirichlet distribution as independent Gamma distributions.

Fixed omission of concentration parameter from Dirichlet Process conditional posterior on groupings.

All functions complete and tested for errors. Need to run for a long time to see if results correct.

Tidied up, and documented DAG and MCMC framework.  model.R contains the SA model, built using the DAG.  test.R contains test functions.

Added AdaptiveLogDirMRW for updating Dirichlet process. Still not perfect, but better than simple MRW.

Converted to R6 classes.  Still having trouble sampling from the Dirichlet distribution with the R's.

Refactored code for graph-based representation of the model.  Now need to write MCMC updaters for Stochastic nodes.

Tidied library dependencies, making use of the "Imports" statement in DESCRIPTION, rather than calls to library(...) all over the place.