-#' @title Forecaster (abstract class)
+#' Forecaster
#'
-#' @description Abstract class to represent a forecaster (they all inherit this)
+#' Forecaster (abstract class, implemented by all forecasters).
#'
-#' @field params List of computed parameters, for post-run analysis (dev)
-#' @field data Dataset, object of class Data
-#' @field pjump Function: how to predict the jump at day interface ?
-Forecaster = setRefClass(
- Class = "Forecaster",
-
- fields = list(
- params = "list",
- data = "Data",
- pjump = "function"
+#' A Forecaster object encapsulates parameters (which can be of various kinds, for
+#' example "Neighbors" method stores informations about the considered neighborhood for
+#' the current prediction task) and one main function: \code{predictSerie()}. This last
+#' function (by default) calls \code{predictShape()} to get a forecast of a centered
+#' serie, and then calls the "jump prediction" function if it's provided -- see "field"
+#' section -- to adjust it based on the last observed values. The main method in derived
+#' forecasters is \code{predictShape()}; see 'Methods' section.
+#'
+#' @usage # Forecaster$new(pjump) #warning: predictShape() is unimplemented
+#'
+#' @field .params List of computed parameters (if applicable).
+#' @field .pjump Function: how to predict the jump at day interface? The arguments of
+#' this function are -- in this order:
+#' \itemize{
+#' \item data: object output of \code{getData()},
+#' \item today: index of the current day in data (known until predict_from-1),
+#' \item memory: number of days to use in the past (including today),
+#' \item predict_from: first time step to predict (in [1,24])
+#' \item horizon: last time step to predict (in [predict_from,24]),
+#' \item params: optimized parameters in the main method \code{predictShape()},
+#' \item ...: additional arguments.
+#' }
+#' .pjump returns an estimation of the jump after the last observed value.
+#'
+#' @section Methods:
+#' \describe{
+#' \item{\code{initialize(data, pjump)}}{
+#' Initialize a Forecaster object with a Data object and a jump prediction function,
+#' or NULL if \code{predictShape()} returns an adjusted curve.}
+#' \item{\code{predictSerie(data,today,memory,predict_from,horizon,...)}}{
+#' Predict the next curve (at index today) from predict_from to horizon (hours), using
+#' \code{memory} days in the past.}
+#' \item{\code{predictShape(data,today,memory,predict_from,horizon,...)}}{
+#' Predict the shape of the next curve (at index today) from predict_from to horizon
+#' (hours), using \code{memory} days in the past.}
+#' \item{\code{getParameters()}}{
+#' Return (internal) parameters.}
+#' }
+#'
+#' @docType class
+#' @format R6 class
+#'
+Forecaster = R6::R6Class("Forecaster",
+ private = list(
+ .params = list(),
+ .pjump = NULL
),
-
- methods = list(
- initialize = function(...)
+ public = list(
+ initialize = function(pjump)
{
- "Initialize (generic) Forecaster object"
-
- callSuper(...)
- if (!hasArg(data))
- stop("Forecaster must be initialized with a Data object")
- params <<- list()
+ private$.pjump <- pjump
+ invisible(self)
},
- predict = function(today, memory, horizon, ...)
+ predictSerie = function(data, today, memory, predict_from, horizon, ...)
{
- "Obtain a new forecasted time-serie"
-
# Parameters (potentially) computed during shape prediction stage
- predicted_shape = predictShape(today, memory, horizon, ...)
- predicted_delta = pjump(data, today, memory, horizon, params, ...)
- # Predicted shape is aligned it on the end of current day + jump
- predicted_shape + tail(data$getSerie(today),1) - predicted_shape[1] + predicted_delta
- },
- predictShape = function(today, memory, horizon, ...)
- {
- "Shape prediction (centered curve)"
+ predicted_shape <- self$predictShape(data,today,memory,predict_from,horizon,...)
+ predicted_delta <-
+ if (is.null(private$.pjump))
+ NULL
+ else
+ private$.pjump(data,today,memory,predict_from,horizon,private$.params,...)
- #empty default implementation: to implement in inherited classes
+ # Predicted shape is aligned on the end of current day + jump (if jump!=NULL)
+ c( data$getSerie(today)[if (predict_from>=2) 1:(predict_from-1) else c()],
+ predicted_shape + ifelse( is.null(private$.pjump),
+ 0,
+ predicted_delta - predicted_shape[1] +
+ ifelse(predict_from>=2,
+ data$getSerie(today)[predict_from-1], tail(data$getSerie(today-1),1)) ) )
},
+ predictShape = function(data, today, memory, predict_from, horizon, ...)
+ NULL #empty default implementation: to implement in inherited classes
+ ,
getParameters = function()
- {
- params
- }
+ private$.params
)
)