group_dt

    描述

    在指定的组内执行数据操作。

    Usage

    group_dt(.data, by = NULL, …)

    rowwise_dt(.data, …)

    Arguments

    .data A data.frame
    by 分组变量的无引号名称列表中的分组变量的无引号名称。
    … 可以在data.frame上实现的任何数据操作参数。

    detail

    如果您想在group_dt中使用summarise_dt和mutate_dt，那么最好在这些函数中使用“by”参数，这样会快得多，因为您不需要使用. sd(这会花费额外的时间来复制)。

    1. iris %>% group_dt(by = Species,slice_dt(1:2)) #equal to
    2. iris %>% group_by_dt(Species) %>% group_exe_dt(head(2))
    4. iris %>% group_dt(Species,filter_dt(Sepal.Length == max(Sepal.Length)))
    5. iris %>% group_dt(Species,summarise_dt(new = max(Sepal.Length)))
    7. # you can pipe in the `group_dt`
    8. iris %>% group_dt(Species,
    9.                   mutate_dt(max= max(Sepal.Length)) %>%
    10.                     summarise_dt(sum=sum(Sepal.Length)))
    12. # for users familiar with data.table, you can work on .SD directly
    13. # following codes get the first and last row from each group
    14. iris %>%
    15.   group_dt(
    16.     by = Species,
    17.     rbind(.SD[1],.SD[.N])
    18.   )
    20. #' # for summarise_dt, you can use "by" to calculate within the group
    21. mtcars %>%
    22.   summarise_dt(
    23.     disp = mean(disp),
    24.     hp = mean(hp),
    25.     by = cyl
    26.   )
    27. # equal to
    28. mtcars %>%
    29.   group_dt(cyl,
    30.            summarise_dt(
    31.              disp = mean(disp),
    32.              hp = mean(hp))
    34.     )
    38. mtcars %>%
    39.   group_dt(by =.(vs,am),
    40.            summarise_dt(avg = mean(mpg)))
    42. mtcars %>%
    43.   group_dt(by =.(vs,am),
    44.            summarise_dt(avg = mean(mpg)))
    47. # examples for `rowwise_dt`
    48. df <- data.table(x = 1:2, y = 3:4, z = 4:5)
    49. df %>% mutate_dt(m = mean(c(x, y, z)))
    50. df %>% rowwise_dt(
    51.   mutate_dt(m = mean(c(x, y, z)))
    52. )