英文:
A function for using in apply() in R
问题
有人知道如何定义一个函数，可以在apply()中使用，返回如下数值吗？data[i, data[i, 300]]。
data[i, 300] 返回我的数据中的列号，我想知道每一行这个列号的内容。
可以使用for循环来解决这个问题，如下所示：
for (i in 1:nrow(data)) {
value <- data[i, data[i, 300]]
}
但是对于1000万个观测来说，这需要太长时间。我正在寻找一种使用apply()的快速方法。
感谢您可以提供的任何帮助。
英文:
Is there anyone know how can I define a function to be used in apply() that return the following value? data[i,data[i,300]].
data[i,300] returns a column number in my data, and I am interested to know the content of this column number for each row.
It's easy to use for loop for addressing this issue as follows:
for (i in 1: nrow(data)) {
value<-data[i,data[i,300]] }
But it takes too much time for 10M observations. I am looking for a fast approach using apply().
I appreciate any help you can provide me.
答案1
得分: 1
对于上述示例数据，假设第300列的值包含了每一行中要检查的不同列号，你可以使用下面的代码来获得期望的输出：
df[cbind(1:nrow(df), df[[300]])]
这段代码会根据每一行的第300列的值，选择相应的列来提取数据，就像你描述的那样。
英文:
I am going to assume that the 300th column has different column numbers for each row and for each row you want to check the value in the column specified by this 300th column.
sample data:
set.seed(123)
df <- data.frame(
c1 = sample(letters, 100, replace = TRUE),
c2 = sample(letters, 100, replace = TRUE),
c3 = sample(letters, 100, replace = TRUE),
c4 = sample(letters, 100, replace = TRUE),
c5 = sample(letters, 100, replace = TRUE),
d = sample(1:5, size = 100, replace = TRUE)
)
> head(df)
c1 c2 c3 c4 c5 d
1 o o p b r 2
2 s u w m l 2
3 n e a x o 3
4 c h h q s 5
5 j s h c u 1
6 r j j r f 3
Thus for above sample, let's assume the 6th column, column d is the one that gives the column numbers, so the expected output should be df[1, 2], df[2, 2], df[3, 3], df[4, 5], df[5, 1], df[6, 3], which evaluates to c('o', 'u','a', 's', 'j', 'j')
To get this we can use:
df[cbind(1:nrow(df), df[[6]])] # change 6 with 300 for your case
通过集体智慧和协作来改善编程学习和解决问题的方式。致力于成为全球开发者共同参与的知识库，让每个人都能够通过互相帮助和分享经验来进步。
评论