英文:
group rows data frame by multiple ranges of same column
问题
Here's the code you provided with the translated parts:
# 给定这些数据:id <- c("1","1", "1","2","2","2","3","3","3","4","4","4","5","5","5","6","6","6")value <- c("1", "2", "3", "4", "5", "6", "7", "8","9","10","11","12","13","14","15","16","17","18")value2 <- c("1", "2", "3", "4", "5", "6", "7", "8","9","10","11","12","13","14","15","16","17","18")value3 <- c("1", "2", "3", "4", "5", "6", "7", "8","9","10","11","12","13","14","15","16","17","18")df <- data.frame(id, value, value2, value3)# 我想按多个范围（group1: 1-2 和 5-6; group2:3-4）对行进行分组，并根据 value 进行汇总，以便最终结果如下所示:newname <- c("newname1", "newname2")sumvalues <- c("114", "57")sumvalues2 <- c("114", "57")sumvalues3 <- c("114", "57")df2 <- data.frame(newname, sumvalues, sumvalues2, sumvalues3)# 当新组（newname）有一个范围时，我已经尝试过以下方法，但我无法弄清如何将多个范围集成到一个新组中。data_values_range <- data_values %>%# 将值聚合到范围中mutate(ranges = cut(group, seq(1, 6, 1))) %>%group_by(ranges) %>%summarize(sumvalues = sum(value)) %>%as.data.frame()data_values_range
Note: I've translated the comments and variable names in the code, but the core code logic remains the same.
英文:
Given this data:
id <- c("1","1", "1","2","2","2","3","3","3","4","4","4","5","5","5","6","6","6")value <- c("1", "2", "3", "4", "5", "6", "7", "8","9","10","11","12","13","14","15","16","17","18")value2 <- c("1", "2", "3", "4", "5", "6", "7", "8","9","10","11","12","13","14","15","16","17","18")value3 <- c("1", "2", "3", "4", "5", "6", "7", "8","9","10","11","12","13","14","15","16","17","18")df <- data.frame(id, value, value2, value3)
I would like to group the rows in two groups by multiple ranges (group1: 1-2 and 5-6; group2:3-4) and summarize by value so that the end result is as follows:
newname <- c("newname1", "newname2")sumvalues <- c("114", "57")sumvalues2 <- c("114", "57")sumvalues3 <- c("114", "57")df2 <- data.frame(newname, sumvalues, sumvalues2, sumvalues3)
I have tried the following when there is one single range of each new group (newname) but I can't figure out how to integrate several ranges into one new group
data_values_range <- data_values %>% # Aggregate values in rangemutate(ranges = cut(group,seq(1, 6, 1))) %>%group_by(ranges) %>%dplyr::summarize(sumvalues = sum(value)) %>%as.data.frame()data_values_range
in the case that there were more than one columns other than id, I would like that the end result shows the sum of the value of those columnes grouped by the new groups
答案1
得分: 1
以下是翻译好的部分：
# 我们可以使用以下代码library(dplyr) # >= 1.1.0df %>%type.convert(as.is = TRUE) %>%group_by(newname = case_match(id, c(1, 2, 5, 6) ~ 'newname1',c(3, 4) ~ 'newname2',.default = 'other')) %>%select(-id) %>%reframe(across(where(is.numeric), ~ sum(.x, na.rm = TRUE),.names = "sum{.col}"))
-output
# 一个 tibble: 2 × 4newname sumvalue sumvalue2 sumvalue3<chr> <int> <int> <int>1 newname1 114 114 1142 newname2 57 57 57
<details><summary>英文:</summary>We could use
library(dplyr)# >= 1.1.0
df %>%
type.convert(as.is = TRUE) %>%
group_by(newname = case_match(id, c(1,2, 5, 6) ~ 'newname1',
c(3, 4)~ 'newname2',
.default = 'other')) %>%
select(-id) %>%
reframe(across(where(is.numeric), ~ sum(.x, na.rm = TRUE),
.names = "sum{.col}"))
-output
A tibble: 2 × 4
newname sumvalue sumvalue2 sumvalue3
<chr> <int> <int> <int>
1 newname1 114 114 114
2 newname2 57 57 57
</details># 答案2**得分**: 0你可以创建一个命名的组列表，然后以长格式获取它们，并将它们与原始的 `df` 连接，以对每个唯一的 `name` 进行求和。```Rlibrary(tidyverse)groups <- list(newname1 = c(1, 2, 5, 6), newname2 = c(3, 4))enframe(groups, value = "new_value") %>%unnest(new_value) %>%inner_join(df, by = c("new_value" = "id"), multiple = "all") %>%summarise(value = sum(value), .by = name)# name value# <chr> <int>#1 newname1 114#2 newname2 57
数据
我不确定为什么数据框 df 中的数字存储为字符。使用 type.convert 将其更改为数字。
df <- type.convert(df, as.is = TRUE)
英文:
You may create a named list of groups that you want to create. Get them in long format and join with original df to sum for each unique name.
library(tidyverse)groups <- list(newname1 = c(1, 2, 5, 6), newname2 = c(3, 4))enframe(groups, value = "new_value") %>%unnest(new_value) %>%inner_join(df, join_by(new_value == id), multiple = "all") %>%summarise(value = sum(value), .by = name)# name value# <chr> <int>#1 newname1 114#2 newname2 57
data
I am not sure why the numbers are stored as characters in the dataframe df. Using type.convert will change them to numbers.
df <- type.convert(df, as.is = TRUE)
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