如何根据在DF2中获得的重要变量对DF1中的列变量进行子集化？ [英] How do I subset column variables in DF1 based on the important variables I got in DF2?

问题描述

我有2个这样的df

ID = c('x1','x2','x5')
df1 <- data.frame(ID)

x1 = c(1,2,3,4,5)
x2 = c(11,12,13,14,15)
x3 = c(21,22,23,24,25)
x4 = c(31,32,33,34,35)
x5 = c(41,42,43,44,45)
df2 <- data.frame(x1,x2,x3,x4,x5)

所需的输出

  x1 x2 x5
1  1 11 41
2  2 12 42
3  3 13 43
4  4 14 44
5  5 15 45

我希望我的新数据集只包含那些在df1中标识为重要的变量（即x1，x2，x5），并带有df2的值。

I would like my new dataset to contain only those variables that are identified in df1 as important (i.e: x1,x2,x5) with the values from df2.

在这个简单的数据集中，我知道可以做到这一点，但只需删除df2中的x3，x4，但理想情况下，我想将其应用于更大的数据集超过100个变量，因此希望以编程方式进行。

In this simple dataset, I know I could do this but just removing x3,x4 in df2 but ideally I would like to apply it to a larger data set where I have more than 100 variables and hence would like to do it programatically.

推荐答案

我找不到重复对象，因此这里只是 as的子集。字符（df1 $ ID）

I can't find a dupe so here goes- simply subset by the values of as.character(df1$ID) as in

df2[as.character(df1$ID)] ## Or just `df2[df1$ID]` if its already a character
#   x1 x2 x5
# 1  1 11 41
# 2  2 12 42
# 3  3 13 43
# 4  4 14 44
# 5  5 15 45

as.character 的原因是为了避免通过 df1 $ ID 底层存储模式（整数）进行子设置，而不是按其级别

The reason for as.character is in order to avoid sub-setting by df1$ID underlying storage mode (integer) rather by it's levels

尽管此问题标有 data.table ，因此我们也可以通过引用（如果我们有 data.table ）来做到这一点-无需转换为 character

Though this question is tagged with data.table, so we could also do this by reference (if we have a data.table)- no need to convert to character

setDT(df2)[, setdiff(names(df2), df1$ID) := NULL]
df2
#    x1 x2 x5
# 1:  1 11 41
# 2:  2 12 42
# 3:  3 13 43
# 4:  4 14 44
# 5:  5 15 45

这篇关于如何根据在DF2中获得的重要变量对DF1中的列变量进行子集化？的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！