使用 R tidyr pivot_wide 从多个列名和值中获取宽格式数据 [英] Using R tidyr pivot_wide to get wide-form data from multiple column names and values

问题描述

如何使用 tidyr pivot_wide 将此数据框从长格式转换为宽形式?我尝试应用文档页面上的示例，但我一定遗漏了一些东西.

How can I use tidyr pivot_wide to convert this data frame from long form to wide form? I tried applying the examples on the docs page, but I must be missing something.

数据框

id <- c(1,1,2,2,3,3)
filename <- c('file1a.txt', 'file1b.txt', 
              'file2a.txt', 'file2b.txt', 
              'file3a.txt', 'file3b.txt')
val <- c(832, 834, 221, 878, 2, 19)

df1 <- data.frame(id, filename, val)
view(df1)

<头>

id	文件名	val
1	file1a.txt	832
1	file1b.txt	834
2	file2a.txt	221
2	file2b.txt	878
3	file3a.txt	2
3	file3b.txt	19

期望输出

<头>

id	文件名1	文件名2	val1	val2
1	file1a.txt	file1b.txt	832	834
2	file2a.txt	file2b.txt	221	878
3	file3a.txt	file3b.txt	2	19

失败的尝试

df_wide <- pivot_wider(data = df1, 
                       id_cols = id, 
                       values_from = c("filename", "val"))
view(df_wide)

<头>

id	文件名_	val_
1	1:2	c(832,834)
2	3:4	c(221,878)
3	5:6	c(2,19)

df_wide <- pivot_wider(data = df1, 
                       id_cols = id, 
                       names_from = c("filename", "val"), 
                       values_from = c("filename", "val"))
view(df_wide)

<头>

id	filename_file1a.txt_832	filename_file1b.txt_834	filename_file2a.txt_221	...等
1	file1a.txt	file1b.txt	不适用	...等
2	不适用	不适用	file2a.txt	...等
3	不适用	不适用	不适用	...等

推荐答案

我们需要一个行序列

library(dplyr)
library(tidyr)
library(data.table)
df1 %>%
    mutate(cn = rowid(id)) %>% 
    pivot_wider(names_from = cn, values_from = c(filename, val), names_sep="")

-输出

# A tibble: 3 x 5
#     id filename1  filename2   val1  val2
#  <dbl> <chr>      <chr>      <dbl> <dbl>
#1     1 file1a.txt file1b.txt   832   834
#2     2 file2a.txt file2b.txt   221   878
#3     3 file3a.txt file3b.txt     2    19

---

或者按row_number

df1 %>%
    group_by(id)
    mutate(cn = row_number()) %>% 
    pivot_wider(names_from = cn, values_from = c(filename, val), names_sep="")

---

如果我们不需要使用%>%，指定data为mutated原始数据集，并增加一列'cn' 基于 'id' 的序列

---

If we need not use %>%, specify the data as the mutated original dataset, with an added column 'cn' based on the sequence of 'id'

pivot_wider(mutate(df1, cn = rowid(id)), 
   names_from = cn, values_from = c(filename, val), names_sep="")

这篇关于使用 R tidyr pivot_wide 从多个列名和值中获取宽格式数据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！