映射R中的数据集名称

由 发布于 applyrcombinations
 收藏

I am trying to analyse the iris data-frame with clustering analysis. I was provided here a way to use Map in R to map the data onto sets of hyperparameter combinations listed by expand.grid and collect all results in one table.

我现在想同时针对数据框的修改版本执行此操作。因此，例如：

acc <- function(x){
  first = sum(x)
  second = sum(x^2)
  return(list(First=first,Second=second))
}
tests <- expand.grid(Clustering_Algorithm=c("ward.D","ward.D2","single","complete","average","mcquitty","median","centroid"),
                     DS=c("iris0","iris1","iris2"))
iris0 <- iris
iris1 <- cbind(log(iris[,1:4]),iris[5])
iris2 <- cbind(sqrt(iris[,1:4]),iris[5])
Table <- Map(function(x, ds){acc(table(ds$Species, cutree(hclust(dist(ds.[,1:4]),method=x),3)))},tests[[1]], tests[[2]])

This fails to run for me with the error "Error in ds$Species : $ operator is invalid for atomic vectors". I've tried writing as.character(tests[[2]]) instead which had the same error message. I've even tried options like ds %>% .[,"Species"]) and ds %>% .[,1:4], in which case I get a different error message: " Error in .[, "Species"] : incorrect number of dimensions".

知道如何解决吗？

编辑：

Just tried to use lapply over the list DS and after putting rest of the above in a function, and that gave me the exact same error message.

评论
# 若曦
# 若曦

The issue is that 'ds' is a string, we need to get the value with mget or get

Map(function(x, ds){
    acc(table(get(ds)$Species, cutree(hclust(dist(get(ds)[,1:4]),method=x),3)))
          }, as.character(tests[[1]]), as.character(tests[[2]]))

It can be made more compact if we use mget

Map(function(x, ds) {
      acc(table(ds$Species, cutree(hclust(dist(ds[, 1:4]), method = x), 3)))
   }, as.character(tests[[1]]), mget(as.character(tests[[2]])))
#$ward.D
#$ward.D$First
#[1] 150

#$ward.D$Second
#[1] 6492


#$ward.D2
#$ward.D2$First
#[1] 150

#$ward.D2$Second
#[1] 6352
#....
点赞
评论