将列表列表转换为数据帧

时间:2017-01-23 10:49:34

标签: r list dataframe nested type-conversion

我有一个名为mylist的嵌套列表,其长度为4。

此列表的每个元素都是一个实验:exp1.1exp1.2exp2.1exp2.2

每个实验都包含四个植物生长阶段的长度(以天为单位)的观察结果:EM-V6 V6-R0 R0-R4R4-R9

每个增长阶段都被组织为一个包含yearmean的数据框。

以下是完整的数据:

mylist=structure(list(exp1.1 = structure(list(`EM-V6` = structure(list(
    year = 2011:2100, mean = c(34, 34, 32, 28, 25, 32, 32, 28, 
    27, 30, 32, 31, 33, 28, 26, 31, 33, 27, 34, 26, 28, 27, 27, 
    30, 29, 31, 34, 30, 26, 31, 33, 33, 27, 30, 28, 32, 31, 29, 
    32, 31, 25, 28, 28, 26, 32, 29, 26, 31, 28, 29, 30, 25, 27, 
    32, 27, 28, 28, 30, 24, 30, 29, 29, 29, 28, 26, 28, 26, 26, 
    28, 31, 30, 27, 26, 28, 25, 24, 24, 30, 27, 26, 26, 27, 26, 
    26, 24, 26, 28, 25, 30, 26)), .Names = c("year", "mean"), row.names = c(NA, 
-90L), class = "data.frame"), `V6-R0` = structure(list(year = 2011:2100, 
    mean = c(30, 33, 33, 32, 29, 30, 32, 31, 32, 30, 33, 30, 
    32, 33, 33, 32, 29, 31, 32, 28, 31, 29, 36, 29, 30, 30, 33, 
    31, 33, 30, 34, 32, 29, 31, 28, 30, 30, 29, 34, 31, 32, 31, 
    30, 28, 32, 29, 29, 32, 29, 28, 29, 29, 32, 31, 27, 32, 29, 
    31, 29, 29, 30, 29, 29, 29, 28, 28, 30, 30, 30, 32, 29, 29, 
    30, 29, 29, 29, 28, 28, 29, 30, 29, 29, 29, 30, 28, 30, 30, 
    29, 29, 29)), .Names = c("year", "mean"), row.names = c(NA, 
-90L), class = "data.frame"), `R0-R4` = structure(list(year = 2011:2100, 
    mean = c(31, 32, 32, 33, 32, 32, 33, 31, 34, 32, 33, 33, 
    32, 31, 33, 31, 32, 32, 32, 30, 32, 31, 34, 30, 31, 32, 34, 
    33, 34, 32, 36, 33, 32, 32, 31, 30, 32, 32, 32, 32, 32, 32, 
    31, 30, 30, 31, 32, 32, 30, 30, 32, 31, 31, 32, 30, 32, 29, 
    32, 31, 30, 32, 30, 30, 31, 32, 30, 31, 30, 31, 32, 31, 31, 
    30, 30, 30, 31, 30, 30, 31, 30, 31, 30, 30, 30, 31, 32, 30, 
    31, 30, 30)), .Names = c("year", "mean"), row.names = c(NA, 
-90L), class = "data.frame"), `R4-R9` = structure(list(year = 2011:2100, 
    mean = c(27, 29, 28, 28, 27, 30, 29, 27, 30, 26, 30, 28, 
    29, 28, 29, 27, 29, 28, 25, 26, 26, 25, 27, 27, 27, 28, 30, 
    28, 29, 27, 29, 28, 29, 28, 26, 26, 28, 28, 30, 28, 27, 25, 
    26, 25, 25, 26, 26, 27, 25, 25, 26, 25, 27, 28, 24, 27, 25, 
    28, 26, 24, 27, 26, 27, 25, 26, 26, 24, 26, 25, 26, 24, 25, 
    25, 26, 26, 25, 25, 25, 25, 25, 26, 25, 25, 25, 25, 26, 26, 
    26, 25, 24)), .Names = c("year", "mean"), row.names = c(NA, 
-90L), class = "data.frame")), .Names = c("EM-V6", "V6-R0", "R0-R4", 
"R4-R9")), exp1.2 = structure(list(`EM-V6` = structure(list(year = 2011:2100, 
    mean = c(34, 34, 32, 28, 25, 32, 32, 28, 27, 30, 32, 31, 
    33, 28, 26, 31, 33, 27, 34, 26, 28, 27, 27, 30, 29, 31, 34, 
    30, 26, 31, 33, 33, 27, 30, 28, 32, 31, 29, 32, 31, 25, 28, 
    28, 26, 32, 29, 26, 31, 28, 29, 30, 25, 27, 32, 27, 28, 28, 
    30, 24, 30, 29, 29, 29, 28, 26, 28, 26, 26, 28, 31, 30, 27, 
    26, 28, 25, 24, 24, 30, 27, 26, 26, 27, 26, 26, 24, 26, 28, 
    25, 30, 26)), .Names = c("year", "mean"), row.names = c(NA, 
-90L), class = "data.frame"), `V6-R0` = structure(list(year = 2011:2100, 
    mean = c(30, 33, 33, 32, 29, 30, 32, 31, 32, 30, 33, 30, 
    32, 33, 33, 32, 29, 31, 32, 28, 31, 29, 36, 29, 30, 30, 33, 
    31, 33, 30, 34, 32, 29, 31, 28, 30, 30, 29, 34, 31, 32, 31, 
    30, 28, 32, 29, 29, 32, 29, 28, 29, 29, 32, 31, 27, 32, 29, 
    31, 29, 29, 30, 29, 29, 29, 28, 28, 30, 30, 30, 32, 29, 29, 
    30, 29, 29, 29, 28, 28, 29, 30, 29, 29, 29, 30, 28, 30, 30, 
    29, 29, 29)), .Names = c("year", "mean"), row.names = c(NA, 
-90L), class = "data.frame"), `R0-R4` = structure(list(year = 2011:2100, 
    mean = c(31, 32, 32, 33, 32, 32, 33, 31, 34, 32, 33, 33, 
    32, 31, 33, 31, 32, 32, 32, 30, 32, 31, 34, 30, 31, 32, 34, 
    33, 34, 32, 36, 33, 32, 32, 31, 30, 32, 32, 32, 32, 32, 32, 
    31, 30, 30, 31, 32, 32, 30, 30, 32, 31, 31, 32, 30, 32, 29, 
    32, 31, 30, 32, 30, 30, 31, 32, 30, 31, 30, 31, 32, 31, 31, 
    30, 30, 30, 31, 30, 30, 31, 30, 31, 30, 30, 30, 31, 32, 30, 
    31, 30, 30)), .Names = c("year", "mean"), row.names = c(NA, 
-90L), class = "data.frame"), `R4-R9` = structure(list(year = 2011:2100, 
    mean = c(27, 29, 28, 28, 27, 30, 29, 27, 30, 26, 30, 28, 
    29, 28, 29, 27, 29, 28, 25, 26, 26, 25, 27, 27, 27, 28, 30, 
    28, 29, 27, 29, 28, 29, 28, 26, 26, 28, 28, 30, 28, 27, 25, 
    26, 25, 25, 26, 26, 27, 25, 25, 26, 25, 27, 28, 24, 27, 25, 
    28, 26, 24, 27, 26, 27, 25, 26, 26, 24, 26, 25, 26, 24, 25, 
    25, 26, 26, 25, 25, 25, 25, 25, 26, 25, 25, 25, 25, 26, 26, 
    26, 25, 24)), .Names = c("year", "mean"), row.names = c(NA, 
-90L), class = "data.frame")), .Names = c("EM-V6", "V6-R0", "R0-R4", 
"R4-R9")), exp2.1 = structure(list(`EM-V6` = structure(list(year = 2011:2100, 
    mean = c(34, 34, 32, 28, 25, 32, 32, 28, 27, 30, 32, 31, 
    33, 28, 26, 31, 33, 27, 34, 26, 28, 27, 27, 30, 29, 31, 34, 
    30, 26, 31, 33, 33, 27, 30, 28, 32, 31, 29, 32, 31, 25, 28, 
    28, 26, 32, 29, 26, 31, 28, 29, 30, 25, 27, 32, 27, 28, 28, 
    30, 24, 30, 29, 29, 29, 28, 26, 28, 26, 26, 28, 31, 30, 27, 
    26, 28, 25, 24, 24, 30, 27, 26, 26, 27, 26, 26, 24, 26, 28, 
    25, 30, 26)), .Names = c("year", "mean"), row.names = c(NA, 
-90L), class = "data.frame"), `V6-R0` = structure(list(year = 2011:2100, 
    mean = c(30, 33, 33, 32, 29, 30, 32, 31, 32, 30, 33, 30, 
    32, 33, 33, 32, 29, 31, 32, 28, 31, 29, 36, 29, 30, 30, 33, 
    31, 33, 30, 34, 32, 29, 31, 28, 30, 30, 29, 34, 31, 32, 31, 
    30, 28, 32, 29, 29, 32, 29, 28, 29, 29, 32, 31, 27, 32, 29, 
    31, 29, 29, 30, 29, 29, 29, 28, 28, 30, 30, 30, 32, 29, 29, 
    30, 29, 29, 29, 28, 28, 29, 30, 29, 29, 29, 30, 28, 30, 30, 
    29, 29, 29)), .Names = c("year", "mean"), row.names = c(NA, 
-90L), class = "data.frame"), `R0-R4` = structure(list(year = 2011:2100, 
    mean = c(31, 32, 32, 33, 32, 32, 33, 31, 34, 32, 33, 33, 
    32, 31, 33, 31, 32, 32, 32, 30, 32, 31, 34, 30, 31, 32, 34, 
    33, 34, 32, 36, 33, 32, 32, 31, 30, 32, 32, 32, 32, 32, 32, 
    31, 30, 30, 31, 32, 32, 30, 30, 32, 31, 31, 32, 30, 32, 29, 
    32, 31, 30, 32, 30, 30, 31, 32, 30, 31, 30, 31, 32, 31, 31, 
    30, 30, 30, 31, 30, 30, 31, 30, 31, 30, 30, 30, 31, 32, 30, 
    31, 30, 30)), .Names = c("year", "mean"), row.names = c(NA, 
-90L), class = "data.frame"), `R4-R9` = structure(list(year = 2011:2100, 
    mean = c(27, 29, 28, 28, 27, 30, 29, 27, 30, 26, 30, 28, 
    29, 28, 29, 27, 29, 28, 25, 26, 26, 25, 27, 27, 27, 28, 30, 
    28, 29, 27, 29, 28, 29, 28, 26, 26, 28, 28, 30, 28, 27, 25, 
    26, 25, 25, 26, 26, 27, 25, 25, 26, 25, 27, 28, 24, 27, 25, 
    28, 26, 24, 27, 26, 27, 25, 26, 26, 24, 26, 25, 26, 24, 25, 
    25, 26, 26, 25, 25, 25, 25, 25, 26, 25, 25, 25, 25, 26, 26, 
    26, 25, 24)), .Names = c("year", "mean"), row.names = c(NA, 
-90L), class = "data.frame")), .Names = c("EM-V6", "V6-R0", "R0-R4", 
"R4-R9")), exp2.2 = structure(list(`EM-V6` = structure(list(year = 2011:2100, 
    mean = c(34, 34, 32, 28, 25, 32, 32, 28, 27, 30, 32, 31, 
    33, 28, 26, 31, 33, 27, 34, 26, 28, 27, 27, 30, 29, 31, 34, 
    30, 26, 31, 33, 33, 27, 30, 28, 32, 31, 29, 32, 31, 25, 28, 
    28, 26, 32, 29, 26, 31, 28, 29, 30, 25, 27, 32, 27, 28, 28, 
    30, 24, 30, 29, 29, 29, 28, 26, 28, 26, 26, 28, 31, 30, 27, 
    26, 28, 25, 24, 24, 30, 27, 26, 26, 27, 26, 26, 24, 26, 28, 
    25, 30, 26)), .Names = c("year", "mean"), row.names = c(NA, 
-90L), class = "data.frame"), `V6-R0` = structure(list(year = 2011:2100, 
    mean = c(30, 33, 33, 32, 29, 30, 32, 31, 32, 30, 33, 30, 
    32, 33, 33, 32, 29, 31, 32, 28, 31, 29, 36, 29, 30, 30, 33, 
    31, 33, 30, 34, 32, 29, 31, 28, 30, 30, 29, 34, 31, 32, 31, 
    30, 28, 32, 29, 29, 32, 29, 28, 29, 29, 32, 31, 27, 32, 29, 
    31, 29, 29, 30, 29, 29, 29, 28, 28, 30, 30, 30, 32, 29, 29, 
    30, 29, 29, 29, 28, 28, 29, 30, 29, 29, 29, 30, 28, 30, 30, 
    29, 29, 29)), .Names = c("year", "mean"), row.names = c(NA, 
-90L), class = "data.frame"), `R0-R4` = structure(list(year = 2011:2100, 
    mean = c(31, 32, 32, 33, 32, 32, 33, 31, 34, 32, 33, 33, 
    32, 31, 33, 31, 32, 32, 32, 30, 32, 31, 34, 30, 31, 32, 34, 
    33, 34, 32, 36, 33, 32, 32, 31, 30, 32, 32, 32, 32, 32, 32, 
    31, 30, 30, 31, 32, 32, 30, 30, 32, 31, 31, 32, 30, 32, 29, 
    32, 31, 30, 32, 30, 30, 31, 32, 30, 31, 30, 31, 32, 31, 31, 
    30, 30, 30, 31, 30, 30, 31, 30, 31, 30, 30, 30, 31, 32, 30, 
    31, 30, 30)), .Names = c("year", "mean"), row.names = c(NA, 
-90L), class = "data.frame"), `R4-R9` = structure(list(year = 2011:2100, 
    mean = c(27, 29, 28, 28, 27, 30, 29, 27, 30, 26, 30, 28, 
    29, 28, 29, 27, 29, 28, 25, 26, 26, 25, 27, 27, 27, 28, 30, 
    28, 29, 27, 29, 28, 29, 28, 26, 26, 28, 28, 30, 28, 27, 25, 
    26, 25, 25, 26, 26, 27, 25, 25, 26, 25, 27, 28, 24, 27, 25, 
    28, 26, 24, 27, 26, 27, 25, 26, 26, 24, 26, 25, 26, 24, 25, 
    25, 26, 26, 25, 25, 25, 25, 25, 26, 25, 25, 25, 25, 26, 26, 
    26, 25, 24)), .Names = c("year", "mean"), row.names = c(NA, 
-90L), class = "data.frame")), .Names = c("EM-V6", "V6-R0", "R0-R4", 
"R4-R9"))), .Names = c("exp1.1", "exp1.2", "exp2.1", "exp2.2"
))

我需要做的是将这个嵌套列表“取消列表”到一个如下所示的数据框:

YEAR   EXP   EM-V6   V6-R0   R0-R4   R4-R9
2011  exp1.1  34      30      31      27
2011  exp1.2  34      30      31      27
2011  exp2.1  34      30      31      27
2011  exp1.1  34      30      31      27

这意味着:

 - first year, first experiment, and growth stages.
 - first year, second experiment and growth stages.
 - first year, third experiment and growth stages
 - first year, fourth experiment and growth stages
 - second year, first experiment and growth stages

等等。

如何执行数据转换?

3 个答案:

答案 0 :(得分:7)

使用rbindlist的{​​{1}} - 套餐两次的替代方案:

data.table

或者一气呵成:

library(data.table)
# bind the dataframes in the 'listed lists' together and include the year with the 'id'-parameter
# the resulting 'data.table's are returned as a list
step1 <- lapply(mylist, rbindlist, id = 'stages')
# bind the resulting list together and include the experiment id
step2 <- rbindlist(step1, id = 'experiment')
# reshape to wide format
dcast(step2, year + experiment ~ stages, value.var = 'mean')

给出:

dcast(rbindlist(lapply(mylist, rbindlist, id = 'stages'), id = 'experiment'),
      year + experiment ~ stages, value.var = 'mean')

答案 1 :(得分:6)

Alternate tidyverse:

library(tidyverse)

map_df(mylist, ~bind_rows(., .id="id"), .id="EXP") %>% 
  spread(id, mean)

答案 2 :(得分:1)

我们可以将tidyverse用于更紧凑和可读的代码

library(dplyr)
library(tidyr)
library(purrr)
res1 <- mylist %>%
            #bind the inner datasets and create an id column
            map(bind_rows, .id = "id") %>%
            #bind the outer datasets and create an EXP column
            bind_rows(.id = "EXP") %>% 
            #reshape to wide format
            spread(id, mean) 

head(res1, 4)
#     EXP year EM-V6 R0-R4 R4-R9 V6-R0
#1 exp1.1 2011    34    31    27    30
#2 exp1.1 2012    34    32    29    33
#3 exp1.1 2013    32    32    28    33
#4 exp1.1 2014    28    33    28    32

或者我们可以通过使用mylist循环lapply来处理此问题,然后通过Map cbind创建一个新列“name”names内部list元素,然后rbind list元素与do.call(rbind,现在做第二个Map根据{{names创建一个新列1}}'mylist',rbind list元素,然后来自reshape的{​​{1}}将其转换为'wide'

base R

注意:未使用外部包(100%res <- do.call(rbind, Map(cbind, lapply(mylist, function(x) do.call(rbind, Map(cbind, x, name = names(x)))), EXP= names(mylist))) res2 <- reshape(res, idvar = c("year", "EXP"), timevar = "name", direction = "wide") row.names(res2) <- NULL head(res2, 4) # year EXP mean.EM-V6 mean.V6-R0 mean.R0-R4 mean.R4-R9 #1 2011 exp1.1 34 30 31 27 #2 2012 exp1.1 34 33 32 29 #3 2013 exp1.1 32 33 32 28 #4 2014 exp1.1 28 32 33 28

或使用base R中的dcast转换为'广泛'格式

reshape2