Commit fdb2904d by Poppy Miller

### Fixed up some documentation

parent 44c53fa1
 ... ... @@ -24,7 +24,8 @@ Imports: cluster, stats, gplots, abind SPIn, grDevices RoxygenNote: 5.0.1 KeepSource: TRUE LazyData: TRUE
 ... ... @@ -11,4 +11,9 @@ export(HaldDP) export(PoisGammaDPUpdate) export(PoissonNode) export(StochasticNode) import(dplyr) importFrom(R6,R6Class) importFrom(SPIn,SPIn) importFrom(grDevices,col2rgb) importFrom(grDevices,colorRampPalette) importFrom(stats,median)
 ... ... @@ -5,7 +5,7 @@ # Copyright: Chris Jewell 2016 # # Purpose: Implements credible interval calculation # ##################################################### #' @importFrom SPIn SPIn ci_chenShao = function(x, alpha) { n <- length(x) sorted <- sort(x) ... ... @@ -27,7 +27,7 @@ ci_chenShao = function(x, alpha) { } return(c( median = median(sorted), median = stats::median(sorted), lower = ci.lower, upper = ci.upper )) ... ... @@ -42,7 +42,7 @@ ci_percentiles <- function(x, alpha) { lower_pos <- round(n * (alpha / 2)) return(c( median = median(x), median = stats::median(x), lower = sorted[lower_pos], upper = sorted[upper_pos] )) ... ... @@ -51,14 +51,14 @@ ci_percentiles <- function(x, alpha) { ci_SPIn <- function(x, alpha) { region <- tryCatch({ SPIn(x, conf = 1 - alpha)\$spin SPIn::SPIn(x, conf = 1 - alpha)\$spin }, error = function(cond) { print("Error calculating SPIn interval.") return(c(NA, NA)) }) return(c( median = median(x), median = stats::median(x), lower = region[1], upper = region[2] )) ... ...
 ... ... @@ -5,11 +5,10 @@ # Copyright: Chris Jewell 2016 # # Purpose: Draws a clustered heatmap # ##################################################### are_colours <- function(object) { sapply(object, function(x) { tryCatch( is.matrix(col2rgb(x)), is.matrix(grDevices::col2rgb(x)), error = function(e) FALSE ) ... ... @@ -41,7 +40,7 @@ clusterHeatMap <- function(object, cols, xnames = 1:length(object), hclust_metho # (when using the default white blue colour scheme), # the higher the dissimilarity between the 2 types (i.e. the less # often two type effects are assigned to the same group in the mcmc) hmcols <- colorRampPalette(cols)(299) hmcols <- grDevices::colorRampPalette(cols)(299) heatmap_data <- as.matrix(disim_clust_g) ... ...
 ... ... @@ -8,19 +8,30 @@ ##################################################### #' Builds the source attribution model. Is not intended to be used by a regular user. Developers only here! #' #' @param y 3D array of [type, time, location] of the number of human cases #' @param R 3D array of normalised relative prevalences for each timepoint [type, source, time] #' @param Time a character vector of timepoint ids matching time dimension in y and R #' @param Location a character vector of location ids matching location dimension in y #' @param Prev a 2D array (matrix) of [source, time]. #' @param a_q concentration parameter for the DP #' @param a_theta shape parameter for the Gamma base distribution for the DP #' @param b_theta rate parameter for the Gamma base distribution for the DP #' @param s vector giving group allocation for each type for the DP #' @param theta vector giving values for each group in the DP #' @param a_r 3D array of [type, src, time] for the hyperprior on the relative prevalences R #' Builds the source attribution model. Is not intended to be used by a regular user. #' Developers only here! #' @return Object of \code{\link{R6Class}}. #' @format \code{\link{R6Class}} object. #' @field y 3D array of [type, time, location] of the number of human cases #' @field X 3D array of the number of positive samples for each type, source and time #' [type, source, time] #' @field R 3D array of normalised relative prevalences for each timepoint #' [type, source, time] #' @field Time a character vector of timepoint ids matching time dimension in y and R #' @field Location a character vector of location ids matching location dimension in y #' @field Sources a character vector of source ids matching the source dimension in X #' @field Type a character vector of type ids matching the type dimension in X #' @field prev a 2D array (matrix) of [source, time]. #' @field a_q concentration parameter for the DP #' @field a_theta shape parameter for the Gamma base distribution for the DP #' @field b_theta rate parameter for the Gamma base distribution for the DP #' @field a_r 3D array of [type, src, time] for the hyperprior on the relative prevalences R #' @field a_alpha 3D array of [source, time, location] for the prior on the alpha parameters #' @field s vector giving initial group allocation for each type for the DP #' @field theta vector giving initial values for each group in the DP #' @field alpha 3D array of [source, time, location] giving initial values for the alpha #' parameters DPModel_impl <- R6::R6Class( "DPModel_impl", public = list( ... ...
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!