r 聚合数据框:一些列未更改，一些列聚合 [英] r aggregate dataframe: some columns unchanged, some columns aggregated

问题描述

我曾尝试查看有关聚合、应用等的手册，但我找不到示例，其中某些列被正在应用的函数跳过，而其他列则按原样复制.示例:

I have tried to look at manuals for aggregate, apply, etc, but I can't find examples where some columns are skipped by the function being applied, while other columns are copied as-is. Example:

> olddf = data.frame(code=c("one","one","two"), val1=c(1,2,3), val2=c(4,5,6), val3=c(7,8,9))
> olddf
  code val1 val2 val3
1  one    1    4    7
2  one    2    5    8
3  two    3    6    9
> 

如何聚合 olddf 以便获得新的数据框，其中:

How do I aggregate olddf so that I get a new data frame where:

• 代码列保持原样，
• val1 被跳过
• val2 通常聚合，例如sum()
• 和一个新列是基于较早的聚合之一创建的......例如新列 = sum(val3)/sum(val2)?

• code column is kept as-is,
• val1 is skipped
• val2 is aggregated normally e.g. sum()
• and a new column is created based on one of the earlier aggregates... e.g. new column = sum(val3)/sum(val2)?

我基本上想要:

> newdf
  code val2 newcol
1  one    9 1.6666
2  two    6 1.5000

我想一步完成，而不是定义一个单独的函数来处理每一列/聚合.这可能吗?

I would like to do this in one step rather than defining a separate function to deal with each column / aggregate. Is this possible?

推荐答案

Try data.table

library(data.table)
setDT(olddf)[, .(val2 = sum(val2),
                 newcol = sum(val3)/sum(val2)), by = code]
#    code val2   newcol
# 1:  one    9 1.666667
# 2:  two    6 1.500000

这篇关于r 聚合数据框:一些列未更改，一些列聚合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！