汇总多组列 [英] Summing Multiple Groups of Columns

问题描述

我遇到的情况是我的数据框包含图像分析结果，其中列是图像中特定类的比例，因此示例数据框class_df看起来像:

I have a situation where my data frame contains the results of image analysis where the columns are the proportion of a particular class present in the image, such that an example dataframe class_df would look like:

id    A    B    C    D    E    F
 1 0.20 0.30 0.10 0.15 0.25 0.00 
 2 0.05 0.10 0.05 0.30 0.10 0.40
 3 0.10 0.10 0.10 0.20 0.20 0.30

这些类中的每一个都属于一个功能组，我想创建一个新列，其中根据这些类计算每个功能组的比例.映射class_fg

Each of these classes belongs to a functional group and I want to create new columns where the proportions of each functional group are calculated from the classes. An example mapping class_fg

class         fg
    A          Z
    B          Z
    C          Z
    D          Y
    E          Y
    F          X

期望的结果将是(添加了一行以显示期望的新列):

and the desired result would be (line added to show the desired new columns):

id    A    B    C    D    E    F |    X    Y    Z
 1 0.20 0.30 0.10 0.15 0.25 0.00 | 0.00 0.40 0.60
 2 0.05 0.10 0.05 0.30 0.10 0.40 | 0.40 0.40 0.20
 3 0.10 0.10 0.10 0.20 0.20 0.30 | 0.30 0.40 0.30

我可以一次使用一个功能组来完成

And I can do it one functional group at a time using

first_fg <- class_fg %>%
  filter(fg == "Z") %>%
  select(class) %>%
  unlist()

class_df <- class_df %>%
  mutate(Z = rowSums(select(., one_of(first_fg))))

当然，有更好的方法可以计算每个功能组的行总和，而不必为每个组重复此代码?也许使用purrr?

Surely there is a better way to do this where I can calculate the row sum for each functional group without having to just repeat this code for each group? Maybe using purrr?

推荐答案

我们可以通过 split'class_df'，用map，select列中的list元素循环class_df'并获取rowSums

We could split the 'class_df' by 'class', loop through the list elements with map, select the columns of 'class_df' and get the rowSums

library(tidyverse)
class_fg %>%
    split(.$fg) %>% 
    map_df(~ class_df %>%
                select(one_of(.x$class)) %>% 
                rowSums) %>%
    bind_cols(class_df, .)
#  id    A   B    C    D    E   F   X   Y   Z
#1  1 0.20 0.3 0.10 0.15 0.25 0.0 0.0 0.4 0.6
#2  2 0.05 0.1 0.05 0.30 0.10 0.4 0.4 0.4 0.2
#3  3 0.10 0.1 0.10 0.20 0.20 0.3 0.3 0.4 0.3

---

或者通过nest进行分组，然后通过map对list元素进行ping操作来进行rowSums

---

Or do a group by nesting, and then do the rowSums by mapping over the list elements

class_fg %>% 
   group_by(fg) %>%
   nest %>%
   mutate(out = map(data, ~  class_df %>%
                               select(one_of(.x$class)) %>% 
                               rowSums)) %>% 
   select(-data)  %>%
   unnest %>% 
   unstack(., out ~ fg) %>% 
   bind_cols(class_df, .)

这篇关于汇总多组列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！