R 分配括号内的变量

发布于04月26日

我有一个小数据框，如下所示.

sad <- structure(list(a = c(1, 2), b = c(3, 4)), class = "data.frame", row.names = c(NA, -2L), class = c("tbl_df", "tbl", 
           "data.frame"))

有人能帮我了解一下下面的操作吗？

 sad[, a := sad[, "Ma"] == "Y"]
# A tibble: 2 x 2
      a     b
  <dbl> <dbl>
1     1     3
2     2     4

当我执行上述操作时，我得到了相同的输出，但不确定这里的目的是什么.有人能帮我吗？

推荐答案

Update:经过研究，我发现:What are the differences between "=" and "<-" assignment operators in R?

@Konrad Rudolf的答案讨论了=和<-之间的区别，有趣的是，用<-代替=会抛出一个错误:

# in contrast to sad[, a = anything_here]

sad[, a <- anything_here]

error: Error in `[.tbl_df`(sad, , a <- anything) : object 'anything' not found`

如果你用<-，也可以用your example

sad[, a <- sad[, "Ma"] == "Y"]

从@Nosredna所提到的网站回答中得出结论:

"谷歌的R风格指南通过禁止在作业(job)中使用"="来简化问题.这是一个不错的 Select ."

So all in all maybe it is an assignement issue.

First answer:

让我们考虑一下:把右边的逗号分解成columns.另一方面，逗号的左侧将对行进行操作:

还可以考虑@onyanbu在 comments 中指出的:

# will not work no column a (a is a object)

sad[, a]

# All rows and the column named "a" -> this will work
sad[, 'a']

# will not work because you want to get all rows and column "Ma" -> there is no colum Ma in sad
sad[, "Ma"]

# here you assign nothing to object a
a = sad[, "Ma"]

# consecutevily you compare a = sad[, "Ma"] (e.g. nothing) to string 'Y'
# will not work because 'Ma' column don't exist.
sad[, "Ma"] == "Y"