#' Parameters for cross-validation are either provided or estimated.
#' Model family can be chosen among "rf", "tree", "ppr" and "knn" for now.
#'
+#' @importFrom FNN knn.reg
+#' @importFrom class knn
+#' @importFrom stats ppr
+#' @importFrom randomForest randomForest
+#' @importFrom rpart rpart
+#' @importFrom caret var_seq
+#'
#' @export
Model <- R6::R6Class("Model",
public = list(
#' @param gmodel Generic model returning a predictive function; chosen
#' automatically given data and target nature if not provided.
#' @param params List of parameters for cross-validation (each defining a model)
- initialize = function(data, target, task, gmodel = NA, params = NA) {
- if (is.na(gmodel)) {
+ initialize = function(data, target, task, gmodel = NULL, params = NULL) {
+ if (is.null(gmodel)) {
# (Generic) model not provided
all_numeric <- is.numeric(as.matrix(data))
if (!all_numeric)
# Numerical data
gmodel = ifelse(task == "regression", "ppr", "knn")
}
- if (is.na(params))
+ if (is.null(params))
# Here, gmodel is a string (= its family),
# because a custom model must be given with its parameters.
- params <- as.list(private$getParams(gmodel, data, target))
+ params <- as.list(private$getParams(gmodel, data, target, task))
private$params <- params
if (is.character(gmodel))
gmodel <- private$getGmodel(gmodel, task)
},
#' @description
#' Returns the model at index "index", trained on dataHO/targetHO.
- #' index is between 1 and self$nmodels.
#' @param dataHO Matrix or data.frame
#' @param targetHO Vector of targets (generally numeric or factor)
#' @param index Index of the model in 1...nmodels
get = function(dataHO, targetHO, index) {
private$gmodel(dataHO, targetHO, private$params[[index]])
+ },
+ #' @description
+ #' Returns the parameter at index "index".
+ #' @param index Index of the model in 1...nmodels
+ getParam = function(index) {
+ private$params[[index]]
}
),
private = list(
# No need to expose model or parameters list
- gmodel = NA,
- params = NA,
+ gmodel = NULL,
+ params = NULL,
# Main function: given a family, return a generic model, which in turn
# will output a predictive model from data + target + params.
getGmodel = function(family, task) {
function(dataHO, targetHO, param) {
require(rpart)
method <- ifelse(task == "classification", "class", "anova")
+ if (is.null(colnames(dataHO)))
+ colnames(dataHO) <- paste0("V", 1:ncol(dataHO))
df <- data.frame(cbind(dataHO, target=targetHO))
- model <- rpart(target ~ ., df, method=method, control=list(cp=param))
- function(X) predict(model, X)
+ model <- rpart::rpart(target ~ ., df, method=method, control=list(cp=param))
+ if (task == "regression")
+ type <- "vector"
+ else {
+ if (is.null(dim(targetHO)))
+ type <- "class"
+ else
+ type <- "prob"
+ }
+ function(X) {
+ if (is.null(colnames(X)))
+ colnames(X) <- paste0("V", 1:ncol(X))
+ predict(model, as.data.frame(X), type=type)
+ }
}
}
else if (family == "rf") {
}
}
else if (family == "knn") {
- function(dataHO, targetHO, param) {
- require(class)
- function(X) class::knn(dataHO, X, cl=targetHO, k=param)
+ if (task == "classification") {
+ function(dataHO, targetHO, param) {
+ require(class)
+ function(X) class::knn(dataHO, X, cl=targetHO, k=param)
+ }
+ }
+ else {
+ function(dataHO, targetHO, param) {
+ require(FNN)
+ function(X) FNN::knn.reg(dataHO, X, y=targetHO, k=param)$pred
+ }
}
}
},
# Return a default list of parameters, given a gmodel family
- getParams = function(family, data, target) {
+ getParams = function(family, data, target, task) {
if (family == "tree") {
# Run rpart once to obtain a CV grid for parameter cp
require(rpart)
df <- data.frame(cbind(data, target=target))
ctrl <- list(
+ cp = 0,
minsplit = 2,
minbucket = 1,
- maxcompete = 0,
- maxsurrogate = 0,
- usesurrogate = 0,
- xval = 0,
- surrogatestyle = 0,
- maxdepth = 30)
- r <- rpart(target ~ ., df, method="class", control=ctrl)
+ xval = 0)
+ method <- ifelse(task == "classification", "class", "anova")
+ r <- rpart(target ~ ., df, method=method, control=ctrl)
cps <- r$cptable[-1,1]
+ if (length(cps) <= 1)
+ stop("No cross-validation possible: select another model")
if (length(cps) <= 11)
return (cps)
step <- (length(cps) - 1) / 10
p <- ncol(data)
# Use caret package to obtain the CV grid of mtry values
require(caret)
- caret::var_seq(p, classification = (task == "classificaton"),
+ caret::var_seq(p, classification = (task == "classification"),
len = min(10, p-1))
}
else if (family == "ppr")