6

我有一个数据集，我想删除在 4 个不同列中具有重复信息的数据行。 

foo<- data.frame(g1 = c("1","0","0","1","1"), v1 = c("7","5","4","4","3"), v2 = c("a","b","x","x","e"), y1 = c("y","c","f","f","w"), y2= c("y","y","y","f","c"), y3 = c("y","c","c","f","w"), y4= c("y","y","f","f","c"), y5=c("y","w","f","f","w"), y6=c("y","c","f","f","w"))

foo 然后看起来像：

  g1 v1 v2 y1 y2 y3 y4 y5 y6
1  1  7  a  y  y  y  y  y  y
2  0  5  b  c  y  c  y  w  c
3  0  4  x  f  y  c  f  f  f
4  1  4  x  f  f  f  f  f  f
5  1  3  e  w  c  w  c  w  w

现在，我想根据 Y1-6 列删除任何具有重复数据的行。因此，如果操作正确，则仅会删除第 4 行和第 1 行，因为所有 Y 变量都完全相同。它是一个多列条件。

我相信我很接近，但它只是无法正常工作。

我已经尝试过：new = foo[!(duplicated(foo[,1:6]))] 考虑使用它会搜索的重复命令并且只找到那些完全匹配的命令？

我考虑过使用带有 & 的条件语句，但也不知道该怎么做。
new = foo[foo$y1==foo$y2|foo$y3|foo$y4|foo$y5|foo$y6]

我想过哪个，但我现在不知所措，迷失了方向。我希望 foo 看起来像：

   g1 v1 v2 y1 y2 y3 y4 y5 y6
2  0  5  b  c  y  c  y  w  c
3  0  4  x  f  y  c  f  f  f
5  1  3  e  w  c  w  c  w  w
4

3 回答 3

10 
> foo[apply(foo[ , paste("y", 1:6, sep = "")], 1,
            FUN = function(x) length(unique(x)) > 1 ), ]
  g1 v1 v2 y1 y2 y3 y4 y5 y6
2  0  5  b  c  y  c  y  w  c
3  0  4  x  f  y  c  f  f  f
5  1  3  e  w  c  w  c  w  w
于 2012-09-14T13:34:45.707 回答
2 
foo[apply(foo, 1, function(x) any(x != x[1])),]
于 2012-09-14T13:27:51.603 回答
1 
> foo[ !rowSums( apply( foo[2:6], 2, "!=", foo[1] ) )==0, ]
  y1 y2 y3 y4 y5 y6
2  c  y  c  y  w  c
3  f  y  c  f  f  f
5  w  c  w  c  w  w

> foo[ ! colSums( apply( foo, 1, duplicated, foo[1] ) ) == 5, ]
  y1 y2 y3 y4 y5 y6
2  c  y  c  y  w  c
3  f  y  c  f  f  f
5  w  c  w  c  w  w
于 2012-09-14T13:29:14.830 回答