我正在尝试获取以下列表:
dim_29 <- list(c(114.49337, 20.29176, 390.74801, 592.75864))
dim_30 <- list(c(112.39257, 19.24136, 388.64721, 594.85944))
dim_31 <- list(c(112.39257, 19.24136, 402.30239, 589.60745))
dim_28 <- list(c(113.44297, 19.24136, 374.99204, 587.50665))
为此:
> dim_28 <- unlist(dim_28)
> dim_28
[1] 113.44297 19.24136 374.99204 587.50665
> dim_28 <- paste(dim_28, sep = " ", collapse = ", ")
> dim_28
[1] "113.44297, 19.24136, 374.99204, 587.50665"
通过使用循环或其他方式自动处理所有列表。我必须写哪个循环来完成此操作,或者应该使用Differentnet函数? 任何帮助都会很棒,谢谢!
答案 0 :(得分:3)
如果您的数据结构正确,则循环很容易。您有4个顺序命名的列表。每个列表包含一个向量。相反,您应该有1个列表,其中包含4个向量:
dim_list = list(
d29 = c(114.49337, 20.29176, 390.74801, 592.75864),
d30 = c(112.39257, 19.24136, 388.64721, 594.85944),
d31 = c(112.39257, 19.24136, 402.30239, 589.60745),
d28 = c(113.44297, 19.24136, 374.99204, 587.50665)
)
lapply(dim_list, paste, collapse = ", ")
# $d29
# [1] "114.49337, 20.29176, 390.74801, 592.75864"
#
# $d30
# [1] "112.39257, 19.24136, 388.64721, 594.85944"
#
# $d31
# [1] "112.39257, 19.24136, 402.30239, 589.60745"
#
# $d28
# [1] "113.44297, 19.24136, 374.99204, 587.50665"
答案 1 :(得分:0)
这是我想出的一种方式。
dim_29 <- list(c(114.49337, 20.29176, 390.74801, 592.75864))
dim_30 <- list(c(112.39257, 19.24136, 388.64721, 594.85944))
dim_31 <- list(c(112.39257, 19.24136, 402.30239, 589.60745))
dim_28 <- list(c(113.44297, 19.24136, 374.99204, 587.50665))
list_group = list(dim_28, dim_29, dim_30, dim_31)
output_dat = list()
for (i in 1:4){
dat =list_group[[i]][[1]]
output_dat[i] = paste0(dat, sep = " ", collapse = ", ")
}
希望您对此有帮助。
答案 2 :(得分:0)
> lapply(mget(ls(pattern = "^dim_\\d+")), function(x) paste(unlist(x), sep = " ", collapse = ", "))
$`dim_28`
[1] "113.44297, 19.24136, 374.99204, 587.50665"
$dim_29
[1] "114.49337, 20.29176, 390.74801, 592.75864"
$dim_30
[1] "112.39257, 19.24136, 388.64721, 594.85944"
$dim_31
[1] "112.39257, 19.24136, 402.30239, 589.60745"