我想从数据框中获取子列表。我一直在找东西。
这不是完整列表，例如map()中带有got_chars的示例，而是数据框中的列表。
最后一件事：不记得我的问题是关于数据帧中的子列表，而不是关于boxplot.stats()。在没有boxplot.stats()的情况下，从数据框中获取列表可能会很有用。
这是我的代表：
library(tidyverse)
library(stats)
library(DBI)
con <- dbConnect(RSQLite::SQLite(), ":memory:")
dbWriteTable(con, "mtcars", mtcars)
data_in<-dbGetQuery(con, "SELECT 'cyl'||cyl as cyl,disp FROM mtcars")
data_in
#> cyl disp
#> 1 cyl6.0 160.0
#> 2 cyl6.0 160.0
#> [...snip..snip ...]
#> 28 cyl4.0 95.1
#> 29 cyl8.0 351.0
#> 30 cyl6.0 145.0
#> 31 cyl8.0 301.0
#> 32 cyl4.0 121.0
boxplot.stats.fct <-function (data, ind, code ) {
r<- boxplot.stats( unlist(data[ind]), coef = 1.5, do.conf = TRUE)
# it works but I would like to do that at the return variable
r$quartile1<-r$stats[2]
r$quartile3<-r$stats[3]
# it doesn't works
r$stats_list=r$stats
r[code]<-as.character(data[code])
return(r)
}
data_transposed <-aggregate (disp ~ cyl, data =data_in, FUN=t)
data_transposed
#> cyl
#> 1 cyl4.0
#> 2 cyl6.0
#> 3 cyl8.0
#> disp
#> 1 108.0, 146.7, 140.8, 78.7, 75.7, 71.1, 120.1, 79.0, 120.3, 95.1, 121.0
#> 2 160.0, 160.0, 258.0, 225.0, 167.6, 167.6, 145.0
#> 3 360.0, 360.0, 275.8, 275.8, 275.8, 472.0, 460.0, 440.0, 318.0, 304.0, 350.0, 400.0, 351.0, 301.0
data.stats.out <- as.data.frame(do.call(rbind,
apply(data_transposed,
MARGIN=1,
FUN=boxplot.stats.fct,"disp","cyl"
)
)
)
data.stats.out
#> stats n conf out
#> 1 71.10, 78.85, 108.00, 120.65, 146.70 11 88.08698, 127.91302
#> 2 145.0, 160.0, 167.6, 196.3, 225.0 7 145.9222, 189.2778 258
#> 3 275.8, 301.0, 350.5, 400.0, 472.0 14 308.695, 392.305
#> quartile1 quartile3 stats_list cyl
#> 1 78.85 108 71.10, 78.85, 108.00, 120.65, 146.70 cyl4.0
#> 2 160 167.6 145.0, 160.0, 167.6, 196.3, 225.0 cyl6.0
#> 3 301 350.5 275.8, 301.0, 350.5, 400.0, 472.0 cyl8.0
它可以工作，但是我已经在函数中用r$stats[2]剪切了数据。
(data.stats.out %>% mutate (toto1 = .$quartile1,
toto2 = .$quartile3
)%>% select (cyl,toto1,toto2))
#> cyl toto1 toto2
#> 1 cyl4.0 78.85 108
#> 2 cyl6.0 160 167.6
#> 3 cyl8.0 301 350.5
无效：列表已合并，第1行等于第2行
(data.stats.out %>% mutate (toto1 = .$stats_list[2],
toto2 = .$stats_list[3]
)%>% select (cyl,toto1,toto2))
#> cyl toto1
#> 1 cyl4.0 145.0, 160.0, 167.6, 196.3, 225.0
#> 2 cyl6.0 145.0, 160.0, 167.6, 196.3, 225.0
#> 3 cyl8.0 145.0, 160.0, 167.6, 196.3, 225.0
#> toto2
#> 1 275.8, 301.0, 350.5, 400.0, 472.0
#> 2 275.8, 301.0, 350.5, 400.0, 472.0
#> 3 275.8, 301.0, 350.5, 400.0, 472.0
dbDisconnect(con)
由reprex package（v0.2.1）于2019-02-08创建
我想在功能之外选择列表的第n个。而且不在函数中。
答案 0 :(得分：0)
我找到了解决问题的方法，在unlist中调用了apply()：）
library(tidyverse)
library(stats)
library(DBI)
con <- dbConnect(RSQLite::SQLite(), ":memory:")
dbWriteTable(con, "mtcars", mtcars)
data_in<-dbGetQuery(con, "SELECT 'cyl'||cyl as cyl,disp FROM mtcars")
#extract
data_in[1:5,]
#> cyl disp
#> 1 cyl6.0 160
#> 2 cyl6.0 160
#> 3 cyl4.0 108
#> 4 cyl6.0 258
#> 5 cyl8.0 360
boxplot.stats.fct <-function (data, ind, code ) {
r<- boxplot.stats( unlist(data[ind]), coef = 1.5, do.conf = TRUE)
# I don't want to preset here
# r$quartile1<-r$stats[2]
# r$quartile3<-r$stats[4]
r[code]<-as.character(data[code])
return(r)
}
data_transposed <-aggregate (disp ~ cyl, data =data_in, FUN=t)
data.stats.out <- as.data.frame(
do.call(rbind,
apply(data_transposed,
MARGIN=1,
FUN=boxplot.stats.fct,"disp","cyl"
)
)
)
D<-data.stats.out
D
#> stats n conf out cyl
#> 1 71.10, 78.85, 108.00, 120.65, 146.70 11 88.08698, 127.91302 cyl4.0
#> 2 145.0, 160.0, 167.6, 196.3, 225.0 7 145.9222, 189.2778 258 cyl6.0
#> 3 275.8, 301.0, 350.5, 400.0, 472.0 14 308.695, 392.305 cyl8.0
unlista <-function (data, code ) {
apply(data[code],MARGIN=1,FUN=unlist)
}
# yes ! what I want to select : only at the end to select the nth element of the nth list here,
# for my exemple 2rd et 4th of stats :)
the_df_list_dynamic_answer<-data.frame(
cyl=unlista(D,"cyl"), # or D["cyl"],
quartile1=unlista(D,"stats")[2,],
quartile3=unlista(D,"stats")[4,]
)
the_df_list_dynamic_answer
#> cyl quartile1 quartile3
#> 1 cyl4.0 78.85 120.65
#> 2 cyl6.0 160.00 196.30
#> 3 cyl8.0 301.00 400.00
dbDisconnect(con)
由reprex package（v0.2.1）于2019-02-12创建