堆棧內存溢出
  簡體   English   中英

使用 group by 過濾行並匹配多列中的不同字符串

[英]Filtering rows with group by and matching different strings in multiple columns

 原文  2022-11-23 10:04:09  0   1    r/ dplyr

問題描述

我有一個包含各種變量的數據框。 我的目標是從根本上過濾掉 match1 和 match2 不符合我的條件但對其他變量進行分組的行。 例如，在下面的數據框中，我想按 ID、名稱、所有變量、時間和 time_unit 進行分組。 我創建了一個 n 列，以更清楚地顯示預期的 output 與數據之間的差異。

當按這些分組時，您可以看到 match1 和 match2 列中存在不正確的重復項，即在行 n = 7,10,11,13，因為 match1 和 match2 列不相同，但其他列相同。 即使其中一個匹配列包含 NA，我也想包括行。 注意：第 9 行我希望 group1 也匹配 group3。

我嘗試了不同的方法，但在網上找不到類似的東西。 我如何匹配但也創建一種關於允許匹配的字典/列表？ 也就是說，如果 match1 包含 group1 並且 match2 包含 group2 但也包含 group3，我想保留行，反之亦然。 我的真實數據框包含更多行，並且我更改了數據。

dataframe：

    ID  name var1     var2 var3 var4 var5 match1 time time_unit match2  n
 1:  1 name1 trt1 Flexible 10.0   40   mg Group1    6     Weeks   <NA>  1
 2:  1 name1 trt2 Flexible   NA   NA <NA> Group1    6     Weeks Group1  2
 3:  1 name1 trt3 Flexible 12.5   50   mg Group1    8     Weeks   <NA>  3
 4:  1 name1 trt1 Flexible 10.0   40   mg Group1    8     Weeks   <NA>  4
 5:  1 name1 trt2 Flexible   NA   NA <NA> Group1    8     Weeks Group1  5
 6:  2 name2 trt4    Fixed 10.0   10   mg Group1    0     weeks Group1  6
 7:  2 name2 trt4    Fixed 10.0   10   mg Group1    0     weeks Group2  7
 8:  2 name2 trt5    Fixed 20.0   20   mg Group1    0     weeks Group1  8
 9:  2 name2 trt5    Fixed 20.0   20   mg Group1    0     weeks Group3  9
10:  2 name2 trt5    Fixed 20.0   20   mg Group1    0     weeks Group2 10
11:  2 name2 trt4    Fixed 10.0   10   mg Group2    0     weeks Group1 11
12:  2 name2 trt4    Fixed 10.0   10   mg Group2    0     weeks Group2 12
13:  2 name2 trt5    Fixed 20.0   20   mg Group2    0     weeks Group1 13
14:  2 name2 trt5    Fixed 20.0   20   mg Group2    0     weeks Group2 14
15:  3 name3 trt6 Flexible 10.0   40   mg Group1    0     weeks Group1 15
16:  3 name3 trt2 Flexible   NA   NA <NA> Group1    0     weeks Group1 16
17:  3 name3 trt6 Flexible 10.0   40   mg Group1    8     Weeks Group1 17
18:  3 name3 trt2 Flexible   NA   NA <NA> Group1    8     weeks Group1 18

預計 output：

    ID  name var1     var2 var3 var4 var5 match1 time time_unit match2  n
 1:  1 name1 trt1 Flexible 10.0   40   mg Group1    6     Weeks   <NA>  1
 2:  1 name1 trt2 Flexible   NA   NA <NA> Group1    6     Weeks Group1  2
 3:  1 name1 trt3 Flexible 12.5   50   mg Group1    8     Weeks   <NA>  3
 4:  1 name1 trt1 Flexible 10.0   40   mg Group1    8     Weeks   <NA>  4
 5:  1 name1 trt2 Flexible   NA   NA <NA> Group1    8     Weeks Group1  5
 6:  2 name2 trt4    Fixed 10.0   10   mg Group1    0     weeks Group1  6
 7:  2 name2 trt5    Fixed 20.0   20   mg Group1    0     weeks Group1  8
 8:  2 name2 trt5    Fixed 20.0   20   mg Group1    0     weeks Group3  9
 9:  2 name2 trt4    Fixed 10.0   10   mg Group2    0     weeks Group2 12
10:  2 name2 trt5    Fixed 20.0   20   mg Group2    0     weeks Group2 14
11:  3 name3 trt6 Flexible 10.0   40   mg Group1    0     weeks Group1 15
12:  3 name3 trt2 Flexible   NA   NA <NA> Group1    0     weeks Group1 16
13:  3 name3 trt6 Flexible 10.0   40   mg Group1    8     weeks Group1 17
14:  3 name3 trt2 Flexible   NA   NA <NA> Group1    8     weeks Group1 18

這是輸出數據：

structure(list(ID = c(1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 
2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 3L), name = c("name1", "name1", 
"name1", "name1", "name1", "name2", "name2", "name2", "name2", 
"name2", "name2", "name2", "name2", "name2", "name3", "name3", 
"name3", "name3", "name3"), var1 = c("trt1", "trt2", "trt3", 
"trt1", "trt2", "trt4", "trt4", "trt5", "trt5", "trt5", "trt4", 
"trt4", "trt5", "trt5", "trt6", "trt2", "trt6", "trt2", "trt6"
), var2 = c("Flexible", "Flexible", "Flexible", "Flexible", "Flexible", 
"Fixed", "Fixed", "Fixed", "Fixed", "Fixed", "Fixed", "Fixed", 
"Fixed", "Fixed", "Flexible", "Flexible", "Flexible", "Flexible", 
"Flexible"), var3 = c(10, NA, 12.5, 10, NA, 10, 10, 20, 20, 20, 
10, 10, 20, 20, 10, NA, 10, NA, 10), var4 = c(40L, NA, 50L, 40L, 
NA, 10L, 10L, 20L, 20L, 20L, 10L, 10L, 20L, 20L, 40L, NA, 40L, 
NA, 40L), var5 = c("mg", NA, "mg", "mg", NA, "mg", "mg", "mg", 
"mg", "mg", "mg", "mg", "mg", "mg", "mg", NA, "mg", NA, "mg"), 
    match1 = c("Group1", "Group1", "Group1", "Group1", "Group1", 
    "Group1", "Group1", "Group1", "Group1", "Group1", "Group2", 
    "Group2", "Group2", "Group2", "Group1", "Group1", "Group1", 
    "Group1", "Group1"), time = c(6L, 6L, 8L, 8L, 8L, 0L, 0L, 
    0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 8L, 8L), time_unit = c("Weeks", 
    "Weeks", "Weeks", "Weeks", "Weeks", "weeks", "weeks", "weeks", 
    "weeks", "weeks", "weeks", "weeks", "weeks", "weeks", "weeks", 
    "weeks", "weeks", "weeks", "Weeks"), match2 = c(NA, "Group1", 
    NA, NA, "Group1", "Group1", "Group2", "Group1", "Group3", 
    "Group2", "Group1", "Group2", "Group1", "Group2", "Group1", 
    "Group1", "Group1", "Group1", "Group1"), n = 1:19), row.names = c(NA, 
-19L), class = c("data.table", "data.frame"), .internal.selfref = <pointer: 0x0000019a9a14a3e0>)

我嘗試了不同的方法，但在網上找不到類似的東西。 我如何匹配但也創建一種關於允許匹配的字典/列表？

1 個解決方案

解決方案1
 1  2022-11-23 10:28:34

我可能會使用 case_when 讓列匹配任意值。 但是，我不確定分組在這里有什么影響，所以我不確定我是否完全理解你想要做什么。 也許這會幫助你開始

library(tidyverse)

df <- tibble(
  ID = c(
    1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L,
    2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 3L
  ), name = c(
    "name1", "name1",
    "name1", "name1", "name1", "name2", "name2", "name2", "name2",
    "name2", "name2", "name2", "name2", "name2", "name3", "name3",
    "name3", "name3", "name3"
  ), var1 = c(
    "trt1", "trt2", "trt3",
    "trt1", "trt2", "trt4", "trt4", "trt5", "trt5", "trt5", "trt4",
    "trt4", "trt5", "trt5", "trt6", "trt2", "trt6", "trt2", "trt6"
  ), var2 = c(
    "Flexible", "Flexible", "Flexible", "Flexible", "Flexible",
    "Fixed", "Fixed", "Fixed", "Fixed", "Fixed", "Fixed", "Fixed",
    "Fixed", "Fixed", "Flexible", "Flexible", "Flexible", "Flexible",
    "Flexible"
  ), var3 = c(
    10, NA, 12.5, 10, NA, 10, 10, 20, 20, 20,
    10, 10, 20, 20, 10, NA, 10, NA, 10
  ), var4 = c(
    40L, NA, 50L, 40L,
    NA, 10L, 10L, 20L, 20L, 20L, 10L, 10L, 20L, 20L, 40L, NA, 40L,
    NA, 40L
  ), var5 = c(
    "mg", NA, "mg", "mg", NA, "mg", "mg", "mg",
    "mg", "mg", "mg", "mg", "mg", "mg", "mg", NA, "mg", NA, "mg"
  ),
  match1 = c(
    "Group1", "Group1", "Group1", "Group1", "Group1",
    "Group1", "Group1", "Group1", "Group1", "Group1", "Group2",
    "Group2", "Group2", "Group2", "Group1", "Group1", "Group1",
    "Group1", "Group1"
  ), time = c(
    6L, 6L, 8L, 8L, 8L, 0L, 0L,
    0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 8L
  ), time_unit = c(
    "Weeks",
    "Weeks", "Weeks", "Weeks", "Weeks", "weeks", "weeks", "weeks",
    "weeks", "weeks", "weeks", "weeks", "weeks", "weeks", "weeks",
    "weeks", "weeks", "weeks", "Weeks"
  ), match2 = c(
    NA, "Group1",
    NA, NA, "Group1", "Group1", "Group2", "Group1", "Group3",
    "Group2", "Group1", "Group2", "Group1", "Group2", "Group1",
    "Group1", "Group1", "Group1", "Group1"
  ), n = 1:19
)

df %>% 
    group_by(across(.cols = c(ID, name, starts_with("var"), time, time_unit))) %>% 
    mutate(custom_filter = case_when(
        match1 == match2 ~ 1,
        is.na(match1) | is.na(match2) ~ 1,
        match1 == "Group1" & match2 == "Group3" ~ 1,
        TRUE ~ 0))
#> # A tibble: 19 x 13
#> # Groups:   ID, name, var1, var2, var3, var4, var5, time, time_unit [10]
#>       ID name  var1  var2     var3  var4 var5  match1  time time_~1 match2     n
#>    <int> <chr> <chr> <chr>   <dbl> <int> <chr> <chr>  <int> <chr>   <chr>  <int>
#>  1     1 name1 trt1  Flexib~  10      40 mg    Group1     6 Weeks   <NA>       1
#>  2     1 name1 trt2  Flexib~  NA      NA <NA>  Group1     6 Weeks   Group1     2
#>  3     1 name1 trt3  Flexib~  12.5    50 mg    Group1     8 Weeks   <NA>       3
#>  4     1 name1 trt1  Flexib~  10      40 mg    Group1     8 Weeks   <NA>       4
#>  5     1 name1 trt2  Flexib~  NA      NA <NA>  Group1     8 Weeks   Group1     5
#>  6     2 name2 trt4  Fixed    10      10 mg    Group1     0 weeks   Group1     6
#>  7     2 name2 trt4  Fixed    10      10 mg    Group1     0 weeks   Group2     7
#>  8     2 name2 trt5  Fixed    20      20 mg    Group1     0 weeks   Group1     8
#>  9     2 name2 trt5  Fixed    20      20 mg    Group1     0 weeks   Group3     9
#> 10     2 name2 trt5  Fixed    20      20 mg    Group1     0 weeks   Group2    10
#> 11     2 name2 trt4  Fixed    10      10 mg    Group2     0 weeks   Group1    11
#> 12     2 name2 trt4  Fixed    10      10 mg    Group2     0 weeks   Group2    12
#> 13     2 name2 trt5  Fixed    20      20 mg    Group2     0 weeks   Group1    13
#> 14     2 name2 trt5  Fixed    20      20 mg    Group2     0 weeks   Group2    14
#> 15     3 name3 trt6  Flexib~  10      40 mg    Group1     0 weeks   Group1    15
#> 16     3 name3 trt2  Flexib~  NA      NA <NA>  Group1     0 weeks   Group1    16
#> 17     3 name3 trt6  Flexib~  10      40 mg    Group1     0 weeks   Group1    17
#> 18     3 name3 trt2  Flexib~  NA      NA <NA>  Group1     0 weeks   Group1    18
#> 19     3 name3 trt6  Flexib~  10      40 mg    Group1     8 Weeks   Group1    19
#> # ... with 1 more variable: `case_when(...)` <dbl>, and abbreviated variable
#> #   name 1: time_unit

創建於 2022-11-23，使用reprex v2.0.2

暫無
暫無

聲明:本站的技術帖子網頁，遵循CC BY-SA 4.0協議，如果您需要轉載，請注明本站網址或者原文地址。任何問題請咨詢:yoyou2525@163.com.

相關問題 通過匹配不同行中兩列的值來過濾數據框 按不同列過濾行 搜索不同列中的匹配行 在 R 中的多行上過濾一個組 匹配字符串在多列上循環 在 R 中跨多個列過濾行 將多列中的字符串拆分為 R 中的多行 匹配來自不同數據幀的信息並過濾掉冗余列 在R中，根據不同列中的字​​符串提取行 根據匹配的字符串返回值，然后擴展到同一組中的其他行
相關標簽
 
粵ICP備18138465號  © 2020-2024 STACKOOM.COM