对于示例数据框:
df1 <- structure(list(country = list("PL", "PL", "PL", "PL", "SE", "SE",
"SE", "SE", "SE", "SE", "SE", "SE"),
region = c("PL42", "PL34",
"PL33", "PL62", "SE22", "SE32", "SE11", "SE31", "SE23", "SE12",
"SE21", "SE33"),
N = c(59L, 55L, 59L, 48L, 233L, 91L, 406L, 148L,
323L, 248L, 163L, 104L),
freq.1 = c(31L, 34L, 37L, 27L, 109L,
53L, 175L, 82L, 169L, 134L, 80L, 51L), result = c(53.52, 61.2,
63.24, 55.7, 46.78, 58.24, 43.1, 55.41, 52.32, 54.03, 49.08,
49.04), level = c(2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2)), .Names = c("country",
"region", "N", "freq.1", "result", "level"), class = c("data.table",
"data.frame"), row.names = c(NA, -12L))
我想应用一个简短的函数来生成一个汇总表:
variable.country <-setDT(df1)[order(country),
list(min_result = min(result),
no.regions =.N,
max_result = max(result),
level= level[1L]),
by = country]
但我收到错误:
Error in forder(x, country) :
Column '1' is type 'list' which is not supported for ordering currently.
我该如何解决这个问题？
最佳答案
“国家”栏是 list .我们可以 unlist列，它应该可以正常工作
df1[, country:= unlist(country)][order(country),
list(min_result = min(result),
no.regions =.N,
max_result = max(result),
level= level[1L]), .(country)]
# country min_result no.regions max_result level
#1: PL 53.52 4 63.24 2
#2: SE 43.10 8 58.24 2
关于r - data.table 中类型为 list 的列的错误消息，我们在Stack Overflow上找到一个类似的问题： https://stackoverflow.com/questions/35778818/