Question

我正在寻找一种方法来提取数据组“ mydf”中因子格式的多个组（“季节”，“ meteo”）的模式（“ meteo2”）。这是我的测试代码，如下所示，但是它不起作用，并导致错误消息。与一组“季节”一起工作。三列均具有“ NA”值。我不确定代码中哪一部分是错误的。任何帮助都非常欢迎！

str(mydf$season)
Factor w/ 4 levels "Spring","Summer",...:
 str(mydf$meteo)
Factor w/ 7 levels "<40","<50","<60",..: 
str(mydf$meteo2)
Factor w/ 4 levels "E","N","S","W": 

# mode function
Mode = function(x){ 
ta = table(x)
tam = max(ta)
if (all(ta == tam))
     mod = NA
else
     if(is.numeric(x))
mod = as.numeric(names(ta)[ta == tam])
else
     mod = names(ta)[ta == tam]
return(mod)}

# extracting mode
dataSummary<-mydf %>% select(season, meteo, meteo2) %>%
mutate(meteo = forcats::fct_explicit_na(meteo)) %>%
group_by(meteo, season) %>%
summarise(m=Mode(meteo2))

dataSummary
error : Column `m` can't promote group 30 to character

这是我的示例数据。

dput(head(mydf_sample))
structure(list(season = structure(c(3L, 3L, 3L, 3L, 3L, 3L), .Label = c("Spring", 
"Summer", "Fall", "Winter"), class = "factor"), meteo2 = structure(c(2L, 
2L, 2L, 1L, 2L, 2L), .Label = c("E", "N", "S", "W"), class = "factor"), 
    meteo = structure(c(6L, 6L, 6L, 6L, 7L, 7L), .Label = c("<40", 
    "<50", "<60", "<70", "<75", "<80", "80+"), class = "factor")), .Names = c("season", 
"meteo2", "meteo"), row.names = c(NA, 6L), class = "data.frame")
>

Answer 1

您的错误未随示例数据一起复制。

但是，如果您的目标是产生模式，则可以通过计算组合并采用最常见的组合来更直接地实现。

mydf %>%
  mutate(meteo = forcats::fct_explicit_na(meteo)) %>%
  count(meteo, season, meteo2) %>%
  arrange(desc(n)) %>%
  distinct(meteo, season, .keep_all = TRUE) %>%
  select(-n)

呼叫distinct将采用它看到的第一个选项，这是最常见的，因为从arrange开始降序。

在平局的情况下，这只会选择选项之一。如果这是一个问题，您可以进行一些调整来选择所有内容。

mydf %>%
  mutate(meteo = forcats::fct_explicit_na(meteo)) %>%
  count(meteo, season, meteo2) %>%
  group_by(meteo, season) %>%
  filter(n == max(n)) %>%
  ungroup() %>%
  select(-n)

Answer 2

根据错误消息，似乎某些组未返回字符值（可能是NA，属于逻辑类）。您可以使用as.character明确地将它们转换为字符。

library(dplyr)

mydf_sample %>% group_by(meteo,season) %>% summarise(m=as.character(Mode(meteo2)))

有什么办法可以通过多组因素来总结？

2 个答案: