upload 0.02

DESCRIPTION

@@ -1,8 +1,8 @@
 Package: outqrf
 Type: Package
 Title: Find the outlier by quantile random forests
-Version: 0.0.1
-Date: 2024-08-22
+Version: 0.0.2
+Date: 2024-08-23
 Author: Lucian Xu
 Maintainer: Lucian Xu <[email protected]>
 Description: Find the outlier by quantile random forests
@@ -15,4 +15,6 @@ Roxygen: list(markdown = TRUE)
 Packaged: 2024-08-22 11:06:44 UTC;
 RoxygenNote: 7.3.2
 Imports: 
-    ranger
+    ranger,
+    dplyr
+URL: https://github.com/flystar233/outqrf

NAMESPACE

@@ -1,6 +1,6 @@
 # Generated by roxygen2: do not edit by hand
 
-export(find_max_index)
+export(find_index)
 export(generateOutliers)
 export(get_quantily_value)
 export(outqrf)

R/outqrf.r

@@ -29,21 +29,57 @@ get_quantily_value <- function(name){
 #' @examples
 #' find_max_index(c(1, 2, 3, 4, 5), 3.5)
 #' @export
-find_max_index <- function(x, y) {
+find_index <- function(x, y) {
     index <- which(x == y)
     if (length(index) >= 1) {
         index_name <- names(x)[index]
         value<-get_quantily_value(index_name)
         return(value)
     } 
     else {
-        closest_index <- which.min(abs(x - y))
+        closest_index <- which(abs(x - y) == min(abs(x-y)))
         closest_index_name <- names(x)[closest_index]
         value <- get_quantily_value(closest_index_name)
         return(value)
     }
 }
 
+#' @title find the right rank
+#' 
+#' @description
+#' This function finds the right rank of a response value in a quantile random forest.
+#' 
+#' @param response a vector of response values
+#' @param outMatrix a matrix of out values
+#' @param median_outMatrix a vector of median out values
+#' @param rmse_ a vector of rmse values
+#' 
+#' @return a vector of ranks
+#' 
+get_right_rank <- function(response,outMatrix,median_outMatrix,rmse_){
+    rank_value <-c()
+    for (i in 1:length(response)){
+        rank_<- find_index(outMatrix[i,],response[i])
+        if (length(rank_)>1){
+            diff = response[i] -median_outMatrix[i]
+            if (abs(diff)>3*rmse_ & diff<0 ){
+                min_value <- min(rank_)
+                rank_value<-c(rank_value,min_value)
+            } else if (abs(diff)>3*rmse_ & diff>0) {
+                max_value <- max(rank_)
+                rank_value<-c(rank_value,max_value)
+            }else {
+                mean_value <- mean(rank_)
+                rank_value<-c(rank_value,mean_value)
+            }       
+        }else {
+             rank_value<-c(rank_value,rank_)
+            }
+
+    }
+    return(rank_value)
+}
+
 #' @title find outliers
 #' 
 #' @description
@@ -101,26 +137,7 @@ outqrf <-function(data,
         diffs = response - median_outMatrix
         rmse_ <- sqrt(sum(diffs*diffs)/(length(diffs)-1))
         rmse <- c(rmse,rmse_)
-        rank_value <-c()
-        for (i in 1:length(response)){
-            rank_<- find_max_index(outMatrix[i,],response[i])
-            if (length(rank_)>1){
-                diff = response[i] -median_outMatrix[i]
-                if (abs(diff)>3*rmse_ & diff<0 ){
-                    min_value <- min(rank_)
-                    rank_value<-c(rank_value,min_value)
-                } else if (abs(diff)>3*rmse_ & diff>0) {
-                    max_value <- max(rank_)
-                    rank_value<-c(rank_value,max_value)
-                }else {
-                    mean_value <- mean(rank_)
-                    rank_value<-c(rank_value,mean_value)
-                }       
-            }else {
-                rank_value<-c(rank_value,rank_)
-                }
-
-        }
+        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)
         outliers <- rbind(outliers,outlier)