如何基于两个重复的行平均行？

Question

我有2000行有一些重复，我想基于重复平均行。

Site  Location Line    Band1
Cal   BC04     BC04A   130
Cal   BC04     BC04B   131
Cal   BC04     BC04C   129

我努力了：

 bind_cols(
    FC %>% distinct(site) %>% .[,-Band1],  # pull out columns we aren't      aggregating
  FC[,c(1, Band1)] %>% group_by(Band1) %>%
    summarise_each(funs(mean)) %>% .[,-1]  # aggregate other columns
)

理想情况下，我想导致：

Site  Location Line    Band1
Cal   BC04     BC04A   130

Answer 1

使用dplyr ，您可以：

df %>%
 group_by(Site) %>%
 filter(n() > 1) %>%
 mutate(Band1 = mean(Band1)) %>%
 slice(1) %>%
 ungroup()

  Site  Location Line  Band1
  <chr> <chr>    <chr> <dbl>
1 Cal   BC04     BC04A   130

这里它保留了重复的“站点”值，计算“Band1”的平均值并选择每个“站点”的第一行。

也许您还想绑定重复和非重复的行：

df %>%
 group_by(Site) %>%
 filter(n() > 1) %>%
 mutate(Band1 = mean(Band1)) %>%
 slice(1) %>%
 ungroup() %>%
 bind_rows(df %>%
            group_by(Site) %>%
            filter(n() < 1) %>%
            ungroup())

或者，如果您只想从每个“站点”的重复值计算它：

df %>%
 group_by(Site, dup = duplicated(Site)) %>%
 filter(dup) %>%
 mutate(Band1 = mean(Band1)) %>%
 slice(1) %>%
 ungroup() %>%
 select(-dup)

  Site  Location Line  Band1
  <chr> <chr>    <chr> <dbl>
1 Cal   BC04     BC04B   130

Answer 2

我喜欢data.table

x <-data.frame( 
Site = c( "Cal","Cal","Cal"),
Location = c( "BC04","BC04","BC04"),
Line = c( "BC04A","BC04B","BC04C"),
Band1= c(130,131, 129))

library( data.table)
x<- data.table( x )

x[ , .(Band1=mean( Band1 )) , by = c("Site","Location")]