Question

如何识别每一行的最大值（例如：b1和a4中的20）是同一行的下一个最大值（b1和a3中的5个）的2倍？

示例输入

name    a1  a2  a3  a4
b1  1   1   5   20
b2  10  9   20  9
b3  1   1   1   1

示例输出

b1  a4  20
b2  a3  20

Answer 1

使用：

library(dplyr)
library(tidyr)
d %>% 
  gather(k, v, -name) %>% 
  arrange(-v, name) %>% 
  group_by(name) %>% 
  mutate(logi = v / lead(v, default = min(v)) >= 2) %>% 
  filter(v == first(v) & logi) %>% 
  select(-logi)

结果是：

# A tibble: 2 x 3
# Groups:   name [2]
    name     k     v
  <fctr> <chr> <int>
1     b1    a4    20
2     b2    a3    20

使用的数据：

d <- structure(list(name = structure(1:3, .Label = c("b1", "b2", "b3"), class = "factor"), 
                    a1 = c(1L, 10L, 1L),
                    a2 = c(1L, 9L, 1L),
                    a3 = c(5L, 20L, 1L),
                    a4 = c(20L, 9L, 1L)),
               .Names = c("name", "a1", "a2", "a3", "a4"), class = "data.frame", row.names = c(NA, -3L))

如何识别比下一个最大值高2倍的最大值

1 个答案: