Question

以下是我的问题的一个例子，以表明我的观点。

 Random <- sample(c("A","B","C","D","E","F","G"), size = 100, replace = 
 TRUE)
 Year <- sample(c(2000,2001,2002,2003,2004,2005), 100, TRUE)
 Value <- sample(c(1,2,3,4), 100, TRUE)

 data <- data.frame(Random,Year,Value)

所以我想要做的是删除＃Table1中一年中不会更改其值的所有行，或者至少只返回＃Table2中的Random列中的行。我在这个例子中标记了我想删除的行，以便更好地理解我的问题。

Answer 1

根据您的逻辑，如果值永远不会改变，则应该将一行作为删除目标。当最小值和最大值相同时，这种情况可以表示为真。试试这个：

df <- data.frame(Random=c("A", "B", "C", "D", "E", "F", "G"),
                 `2000`=c(1,1,0,2,2,0,3),
                 `2001`=c(0,1,0,2,3,0,3),
                 `2002`=c(2,1,0,2,0,1,3),
                 `2003`=c(1,1,0,2,0,0,3),
                 `2004`=c(4,1,0,2,1,0,3),
                 `2005`=c(5,1,0,2,1,0,3), stringsAsFactors=FALSE)

df.target <- df[, !(names(df) %in% c("Random"))]
df[apply(df.target, 1, function(x) min(x)!=max(x)), ]

  Random X2000 X2001 X2002 X2003 X2004 X2005
1      A     1     0     2     1     4     5
5      E     2     3     0     0     1     1
6      F     0     0     1     0     0     0

Demo

修改

如果您还想删除表1中的行，这些行的名称与第二个表中要删除的行相匹配，您可以尝试：

names.rm <- df$Random[apply(df.target, 1, function(x) min(x)==max(x))] table1[!table1$Random %in% names.rm, ]

Answer 2

在@TimBiegeleisen的答案中使用数据集，以下内容也只保留至少有一个不同值的行。

df[apply(df[-1], 1, function(x) any(x[-1] != x[1])), ]

Answer 3

以下是使用rowMins/rowMaxs

的选项

library(matrixStats)
df[rowMins(as.matrix(df[-1])) != rowMaxs(as.matrix(df[-1])),]

或pmin/pmax

df[do.call(pmin, df[-1]) != do.call(pmax, df[-1]),]
#    Random X2000 X2001 X2002 X2003 X2004 X2005
#1      A     1     0     2     1     4     5
#5      E     2     3     0     0     1     1
#6      F     0     0     1     0     0     0

Answer 4

另一种方法是以长格式处理数据，然后重新整形。这是一种更好的方法，因为在data.frames上执行逐行操作的成本很高。这是使用Tim数据集的修改版本的基本R解决方案。

reshape(dat[ave(dat$count, dat$Random, FUN=var) != 0, ],
        direction="wide", idvar="Random", timevar="year")

此处，ave(dat$count, dat$Random, FUN=var)选择具有非零方差的dat$Random值，返回TRUE。生成的向量用于对data.frame进行子集化，然后使用基本R的reshape函数将其重新整形为所需的格式。

返回

  Random count.2000 count.2001 count.2002 count.2003 count.2004 count.2005
1      A          1          0          2          1          4          5
5      E          2          3          0          0          1          1
6      F          0          0          1          0          0          0

数据

dat <- structure(list(Random = c("A", "B", "C", "D", "E", "F", "G", "A", "B", "C", "D", "E", "F", "G", "A", "B", "C", "D", "E", "F", "G", "A", "B", "C", "D", "E", "F", "G", "A", "B", "C", "D", "E", "F", "G", "A", "B", "C", "D", "E", "F", "G"), year = c(2000, 2000, 2000, 2000, 2000, 2000, 2000, 2001, 2001, 2001, 2001, 2001, 2001, 2001, 2002, 2002, 2002, 2002, 2002, 2002, 2002, 2003, 2003, 2003, 2003, 2003, 2003, 2003, 2004, 2004, 2004, 2004, 2004, 2004, 2004, 2005, 2005, 2005, 2005, 2005, 2005, 2005), count = c(1, 1, 0, 2, 2, 0, 3, 0, 1, 0, 2, 3, 0, 3, 2, 1, 0, 2, 0, 1, 3, 1, 1, 0, 2, 0, 0, 3, 4, 1, 0, 2, 1, 0, 3, 5, 1, 0, 2, 1, 0, 3)), .Names = c("Random", "year", "count"), row.names = c(NA, -42L), class = "data.frame")

如果完整行包含零，则删除某些行

4 个答案:

Demo