从其他多列的值有条件地替换多列的值答案

【问题标题】：Conditionally replace values of multiple columns, from values of other multiple columns从其他多列的值有条件地替换多列的值
【发布时间】：2020-02-20 20:28:26
【问题描述】：

假设我有这个数据集：

set.seed (1234); 
data.frame(cbind(a=rep(c("si","no"),30),b=rnorm(60)),
           c=rep(c("d","e","f"),20))  %>% head()

然后我想添加许多列（在本例中我只添加了两列），以识别每个组之间的不同情况（在本例中为“a”列）。

set.seed(1234); 
data.frame(cbind(a=rep(c("si","no"),30),b=rnorm(60)),c=rep(c("d","e","f"),20)) %>% 
group_by(a) %>% dplyr::mutate_at(vars(c(b,c)), .funs= list(dups_hash_ing= ~n_distinct(.)))

此代码留下以下数据集：

如果我用dput设置数据集，结果是

structure(list(a = structure(c(2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 
2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 
2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 
2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 2L, 1L, 
2L, 1L, 2L, 1L), .Label = c("no", "si"), class = "factor"), b = structure(c(22L, 
1L, 51L, 34L, 50L, 57L, 53L, 10L, 47L, 3L, 11L, 23L, 15L, 38L, 
58L, 39L, 41L, 17L, 28L, 21L, 37L, 45L, 29L, 46L, 32L, 48L, 56L, 
52L, 26L, 19L, 35L, 8L, 55L, 20L, 9L, 36L, 2L, 12L, 6L, 42L, 
49L, 43L, 59L, 54L, 31L, 13L, 60L, 44L, 14L, 30L, 7L, 5L, 16L, 
27L, 33L, 18L, 24L, 4L, 25L, 40L), .Label = c("-0.0997905884418961", 
"-0.151736536534977", "-0.198416273822079", "-0.254874652654534", 
"-0.274704218225806", "-0.304721068966714", "-0.324393300483657", 
"-0.400235237343163", "-0.415751788401515", "-0.50873701541522", 
"-0.538070788884863", "-0.60615111526422", "-0.659770093821306", 
"-0.684320344136007", "-0.789646852263761", "-0.933503340589868", 
"-0.965903210133575", "-1.07754212275943", "-1.11444896479736", 
"-1.60708093984972", "-2.07823754188738", "-2.7322195229558", 
"-2.85575865501923", "-3.23315213292314", "0.0295178303214797", 
"0.0326639575014441", "0.116845344986082", "0.162654708118265", 
"0.185513915583057", "0.186492083080971", "0.287709728313787", 
"0.311681028661359", "0.319160238648117", "0.413868915451097", 
"0.418057822385083", "0.42200837321742", "0.485226820569252", 
"0.487814635163685", "0.500694614280786", "0.594273774110513", 
"0.62021020366732", "0.629536099884472", "0.660212631820405", 
"0.677415500438328", "0.696768778564913", "0.700733515544461", 
"0.704180178465512", "0.760462361967838", "0.895171980275539", 
"0.912322161610113", "0.976031734922396", "1.1123628412626", 
"1.16910851401363", "1.17349757263239", "1.49349310261748", "1.84246362620766", 
"1.98373220068438", "2.16803253951933", "2.27348352044748", "2.91914013071762"
), class = "factor"), c = structure(c(1L, 2L, 3L, 1L, 2L, 3L, 
1L, 2L, 3L, 1L, 2L, 3L, 1L, 2L, 3L, 1L, 2L, 3L, 1L, 2L, 3L, 1L, 
2L, 3L, 1L, 2L, 3L, 1L, 2L, 3L, 1L, 2L, 3L, 1L, 2L, 3L, 1L, 2L, 
3L, 1L, 2L, 3L, 1L, 2L, 3L, 1L, 2L, 3L, 1L, 2L, 3L, 1L, 2L, 3L, 
1L, 2L, 3L, 1L, 2L, 3L), .Label = c("d", "e", "f"), class = "factor"), 
    a_dups_hash_ing = c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L, 1L, 1L, 1L), b_dups_hash_ing = c(30L, 30L, 30L, 30L, 
    30L, 30L, 30L, 30L, 30L, 30L, 30L, 30L, 30L, 30L, 30L, 30L, 
    30L, 30L, 30L, 30L, 30L, 30L, 30L, 30L, 30L, 30L, 30L, 30L, 
    30L, 30L, 30L, 30L, 30L, 30L, 30L, 30L, 30L, 30L, 30L, 30L, 
    30L, 30L, 30L, 30L, 30L, 30L, 30L, 30L, 30L, 30L, 30L, 30L, 
    30L, 30L, 30L, 30L, 30L, 30L, 30L, 30L), c_dups_hash_ing = c(3L, 
    3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 
    3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 
    3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 
    3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L)), class = c("grouped_df", 
"tbl_df", "tbl", "data.frame"), row.names = c(NA, -60L), groups = structure(list(
    a = structure(1:2, .Label = c("no", "si"), class = "factor"), 
    .rows = list(c(2L, 4L, 6L, 8L, 10L, 12L, 14L, 16L, 18L, 20L, 
    22L, 24L, 26L, 28L, 30L, 32L, 34L, 36L, 38L, 40L, 42L, 44L, 
    46L, 48L, 50L, 52L, 54L, 56L, 58L, 60L), c(1L, 3L, 5L, 7L, 
    9L, 11L, 13L, 15L, 17L, 19L, 21L, 23L, 25L, 27L, 29L, 31L, 
    33L, 35L, 37L, 39L, 41L, 43L, 45L, 47L, 49L, 51L, 53L, 55L, 
    57L, 59L))), row.names = c(NA, -2L), class = c("tbl_df", 
"tbl", "data.frame"), .drop = TRUE))

我需要做的是逐列替换，如果每组不同案例的数量超过一个，则使用原始列的值。我必须这样做超过 50 列。将仅针对具有 mutate 的一列提供这样的示例：

  dplyr::mutate(b_dups_hash_ing= ifelse(>1,b,0))

我需要为许多变量重复上面提供的代码。这与mutate_at 非常相似（括号中的词是我会做的）。以下示例行不通，但这是我在理想世界中会做的事情，只是为了让您更好地理解我的问题。

dplyr::mutate_at(vars(contains('_dups_hash_ing')), .funs = list(~ifelse(.>1,vars([original]),0)))

【问题讨论】：

请使用set.seed([some number here])设置种子，使其可重现
感谢您的回复。但是，该示例的输出不相关。我对可以替换列中的值的函数特别感兴趣。我会尽力编辑问题，我不知道该怎么做。
@JasonAizkalns 我按照你的要求做了。我添加了种子。谢谢。
尝试使用dput(dataset)，其中dataset 可能是dataframe。
那不是我，只是想通过建议使问题更易于理解。

标签： r replace at-command dplyr

【解决方案1】：

这是你要找的吗？

df %>% dplyr::mutate_at(vars(contains('_dups_hash_ing')), ~ ifelse(. > 1, ., 0)) %>% head
#> # A tibble: 6 x 6
#> # Groups:   a [2]
#>   a     b                  c     a_dups_hash_ing b_dups_hash_ing c_dups_hash_ing
#>   <fct> <fct>              <fct>           <dbl>           <int>           <int>
#> 1 si    -2.7322195229558   d                   0              30               3
#> 2 no    -0.09979058844189… e                   0              30               3
#> 3 si    0.976031734922396  f                   0              30               3
#> 4 no    0.413868915451097  d                   0              30               3
#> 5 si    0.912322161610113  e                   0              30               3
#> 6 no    1.98373220068438   f                   0              30               3

【讨论】：

谢谢，但我需要做这样的事情：df %>% dplyr::mutate(b_dups_hash_ing=ifelse(. > 1,b, 0)) %>% head，但要超过 50 列（我想通过函数 mutate_at 来做到这一点，但我无法想象怎么做）。
是否有一个谓词函数可以传递给 mutate_if 以选择您想要的列，或者您可以简单地将位置的数字向量传递给 mutate_at(vars())？
我可以简单地传递一个位置的数字向量。我有一个列名向量，它们是原始列名 (c(b,c))，以及我创建的列名向量应该具有相同的长度，这些列名包含后缀“_dup_hash_ing”(@ 987654325@)。这就是我在上一步中所做的：dplyr::mutate_at(vars(c(b,c)), .funs= list(dups_hash_ing= ~n_distinct(.)))。现在我可以检测每组是否有多个案例（在 a 列中），如果每组有多个案例，我可以将值替换为原始值。
对不起。我英语说得不太好。我需要用原始变量（“b”和“c”）替换包含“_dup_hash_ing”（“b_dup_hash_ing”和“c_dup_hash_ing”）的变量中的值，如果“_dup_hash_ing”的值大于1 .
我需要一个可以执行以下操作的函数：dplyr::mutate_at(vars(c(b,c)), .funs= funs(dups_hash_ing=ifelse(n_distinct(.)>1, as.character(.),""))) 但通过给我每行不同的值，而不是按组 (a) (group_by(a))