nestedfield on the particle filter class has been split into two logical fields:
particle_deterministicand multistage models (#177)
mcstate::pmcmc_chains_*to always use a file for communication, making them easier to understand and more robust (#179)
mcstate::pmcmc_chains_cleanupfor removing files created by the above, and
mcstate::pmcmc_chains_collectfor automating collecting samples
modelas an argument, more in keeping with other functions.
mcstate::particle_deterministic_stateobject for advanced use of the deterministic particle
mcstate::particle_filter_initialfor creating particle filter initial state functions from restart data.
mcstate::pmcmcwhich can use it to stop calculating a likelood that would be rejected. Primarily useful when running with relatively low numbers of particles and a high variance in the estimator (#138)
pmcmc_chains_runwhich can be used to manually schedule chains over different computing resourcess (#129)
rerun_everyis specified, a new control parameter
rerun_controlcan be used to make this stochastic rerun
pmcmc_controlfor controlling the ratio of fixed:varied steps for nested pMCMC
mcstate::array_flattenfor unshaping an array
pmcmc(these were deprecated in 0.3.0) (#114)
pmcmc_parameters_nestedfor discrete and bounded parameters.
pmcmc_varied_parameterfor parameters that can vary between different populations.
pmcmc_parameters_nestedto hold parameters that vary between populations (
pmcmc_varied_parameter) and parameters that are the same (fixed) between populations (
$begin_runmethod on the particle filter (#78)
pmcmc_parametersobjects for fixing the value for a subset of parameters before running with
prev_stateargument and now use just the current model state. This requires that models compute things like “daily incidence” within model code but simplifies use with irregular time series (#94)
pmcmccan returned sample restart state using the
mcstate::pmcmc_controlwhich can be used to restart the pMCMC part way along the time series (see
pmcmcis now controllable via a new
pmcmccan run chains in parallel using
callr, by specifying
n_workers = nfor
ngreater than 1.