proposed_step_mods: Fit the chunk-level models to a time series, given a set of...

In weecology/LDATS: Latent Dirichlet Allocation Coupled with Time Series Analyses

Description

This function wraps around TS_memo (optionally memoised multinom_TS) to provide a simpler interface within the ptMCMC algorithm and is implemented within propose_step.

Usage

proposed_step_mods(prop_changepts, inputs)

Arguments

prop_changepts	matrix of proposed change points across chains.
inputs	Class ptMCMC_inputs list, containing the static inputs for use within the ptMCMC algorithm.

Value

List of models associated with the proposed step, with an element for each chain.

Examples

  data(rodents)
  document_term_table <- rodents$document_term_table
  document_covariate_table <- rodents$document_covariate_table
  LDA_models <- LDA_set(document_term_table, topics = 2)[[1]]
  data <- document_covariate_table
  data$gamma <- LDA_models@gamma
  weights <- document_weights(document_term_table)
  data <- data[order(data[,"newmoon"]), ]
  saves <- prep_saves(1, TS_control())
  inputs <- prep_ptMCMC_inputs(data, gamma ~ 1, 1, "newmoon", weights,
                               TS_control())
  cpts <- prep_cpts(data, gamma ~ 1, 1, "newmoon", weights, TS_control())
  i <- 1
  pdist <- inputs$pdist
  ntemps <- length(inputs$temps)
  selection <- cbind(pdist$which_steps[i, ], 1:ntemps)
  prop_changepts <- cpts$changepts
  curr_changepts_s <- cpts$changepts[selection]
  prop_changepts_s <- curr_changepts_s + pdist$steps[i, ]
  if(all(is.na(prop_changepts_s))){
    prop_changepts_s <- NULL
  }
  prop_changepts[selection] <- prop_changepts_s
  mods <- proposed_step_mods(prop_changepts, inputs)

weecology/LDATS documentation built on March 28, 2020, 11:20 a.m.