Merge pull request #1 from flystar233/dev

Dev

DESCRIPTION

@@ -10,7 +10,7 @@ Authors@R:
            family = "Xu",
            role = c("aut", "cre"),
            email = "flystar233@gmail.com")
-Description: Provides a method to find the outlier in custom data by quantile random forests("Quantile Regression Forests", Journal of Machine Learning Research, 7(Jun), 983-999, 2006.). It directly calls the ranger function of the ranger package to perform data fitting and prediction. We also implement the evaluation of outlier prediction results.Compared with random forest detection of outliers, this method has higher accuracy and stability.
+Description: Provides a method to find the outlier in custom data by quantile random forests method. Introduced by Meinshausen Nicolai (2006) <doi:10.5555/1248547.1248582>. It directly calls the ranger() function of the 'ranger' package to perform data fitting and prediction. We also implement the evaluation of outlier prediction results. Compared with random forest detection of outliers, this method has higher accuracy and stability on large datasets.
 LazyData: false
 License: MIT + file LICENSE
 Depends: 

R/generateOutliers.r

@@ -8,13 +8,13 @@
 #' @returns data with some outliers.
 #' @export
 #' @examples
-#' generateOutliers(iris, p = 0.05, sd_factor = 5, seed = 123)
+#' generateOutliers(iris, p = 0.05, sd_factor = 5)
 generateOutliers <- function(data, p = 0.05, sd_factor = 5, seed = NULL){
     if (p < 0 || p > 1|| sd_factor <= 0) {
         stop("p and sd_factor must be between 0 and 1")
     }
     if (is.null(seed)) {
-        set.seed(123)
+        set.seed(as.numeric(Sys.time()))
     } else {
         set.seed(seed)
     }

R/outqrf.r

@@ -87,10 +87,11 @@ get_right_rank <- function(response,outMatrix,median_outMatrix,rmse_){
 #' This function finds outliers in a dataset using quantile random forests.
 #'
 #' @param data a data frame
-#' @param quantiles_type '1000':seq(from = 0.001, to = 0.999, by = 0.001), '400':seq(0.0025,0.9975,0.0025)
+#' @param quantiles_type specify the type of quantile generation.Default is 1000.
 #' @param threshold a threshold for outlier detection
 #' @param verbose a boolean value indicating whether to print verbose output
 #' @param impute a boolean value indicating whether to impute missing values
+#' @param weight a boolean value indicating whether to use weight. if TRUE, The actual threshold will be threshold*r2.
 #' @param ... additional arguments passed to the ranger function
 #' @return
 #' An object of class "outqrf" and a list with the following elements.
@@ -119,6 +120,7 @@ outqrf <-function(data,
                   threshold =0.025,
                   impute = TRUE,
                   verbose = 1,
+                  weight = FALSE,
                   ...){
     # Initial check
     if (!is.data.frame(data)) {
@@ -191,7 +193,11 @@ outqrf <-function(data,
         rmse <- c(rmse,rmse_)
         rank_value <- get_right_rank(response,outMatrix,median_outMatrix,rmse_)
         outlier <- data.frame(row = as.numeric(row.names(data)),col = v,observed = response, predicted = median_outMatrix,rank = rank_value)
-        outlier<- outlier|>dplyr::filter(rank<=threshold_low| rank>=threshold_high)
+        if (weight){
+            outlier<- outlier|>dplyr::filter(rank<=threshold_low*qrf$r.squared| rank>=1-threshold_low*qrf$r.squared)}
+        else{
+            outlier<- outlier|>dplyr::filter(rank<=threshold_low| rank>=threshold_high)
+        }
         outliers <- rbind(outliers,outlier)
     }
     # names of the variables

docs/pkgdown.yml

@@ -3,7 +3,7 @@ pkgdown: 2.1.0
 pkgdown_sha: ~
 articles:
   outqrf: outqrf.html
-last_built: 2024-08-28T05:43Z
+last_built: 2024-09-04T09:48Z
 urls:
   reference: flystar233.github.io/outqrf/reference
   article: flystar233.github.io/outqrf/articles