Use loop to create lists inside another list

Question

I would like to create two lists inside another list. Normally I will have to create 50 each one for a state of USA and I'm looking for a way to make it faster.

State <- c("ALABAMA", "ALABAMA", "ALASKA", "ALASKA")
Num <- c(5, 6, 7, 8)
d <- data.frame(State, Num)

uni<-unique(d$State)

data = list(
  for(i in 1 : length(uni)){
    list[[i]](
      name = un[i],
      y = summarise(sum(d$Num[D$State==uni[i]])),
      drilldown = tolower(un[i])
    )
  }
)

More specifically I want each list to include the name of the State in Capital letters, y as the sum of the Num and drilldown as the name of the State in non capital letters. The result should be something like:

name="ALABAMA"
y= 11
drilldown="alabama"

name="ALASKA"
y= 15
drilldown="alaska"

Manually it would be like :

data = list(

      list(
        name = "ALABAMA",
        y = 11,
        drilldown = "alabama"
      ),
      list(
        name = "ALASKA",
        y = 15,
        drilldown = "alaska"
      )

    )

which gives a list of 2 lists of 3 objects.

It may be possible without for()so Im open to other suggestions

Darren Tsai · Accepted Answer · 2019-02-24 17:31:54Z

3

State <- c("ALABAMA", "ALABAMA", "ALASKA", "ALASKA")
Num <- c(5, 6, 7, 8)
d <- data.frame(State, Num, stringsAsFactors = F)

Solution 1 : Use built-in packages

df1 <- within(aggregate(Num ~ State, d, FUN = sum), drilldown <- tolower(State))
output1 <- lapply(split(df1, 1:nrow(df1)), c)
str(output1)

# List of 2
#  $ 1:List of 3
#   ..$ State    : chr "ALABAMA"
#   ..$ Num      : num 11
#   ..$ drilldown: chr "alabama"
#  $ 2:List of 3
#   ..$ State    : chr "ALASKA"
#   ..$ Num      : num 15
#   ..$ drilldown: chr "alaska"

Solution 2 : library(tidyverse)

output2 <- d %>% group_by(State) %>%
                 summarise(Num = sum(Num)) %>%
                 mutate(drilldown = tolower(State)) %>% 
                 transpose  # purrr::transpose
str(output2)

# List of 2
#  $ :List of 3
#   ..$ State    : chr "ALABAMA"
#   ..$ Num      : num 11
#   ..$ drilldown: chr "alabama"
#  $ :List of 3
#   ..$ State    : chr "ALASKA"
#   ..$ Num      : num 15
#   ..$ drilldown: chr "alaska"

edited Feb 24, 2019 at 17:31

answered Feb 24, 2019 at 15:36

Darren Tsai

36.6k6 gold badges27 silver badges58 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

firmo23 Over a year ago

I made an edit. Yous solution gives 1 list with 2 'list of 3' while I would like 2 lists.

Darren Tsai Over a year ago

@firmo23 I don't understand the difference between the output you edited and mine.

Rich Scriven · Accepted Answer · 2019-02-24 17:46:21Z

3

You can run a quick rowsum() to get the group sums, then Map() to assemble the list from its parts.

xx <- with(d, rowsum(Num, State))
Map(list, name=rownames(xx), y=xx, drilldown=tolower(rownames(xx)), USE.NAMES=FALSE)
# [[1]]
# [[1]]$name
# [1] "ALABAMA"
#
# [[1]]$y
# [1] 11
#
# [[1]]$drilldown
# [1] "alabama"
#
#
# [[2]]
# [[2]]$name
# [1] "ALASKA"
#
# [[2]]$y
# [1] 15
#
# [[2]]$drilldown
# [1] "alaska"

edited Feb 24, 2019 at 17:46

answered Feb 24, 2019 at 17:29

Rich Scriven

99.7k11 gold badges190 silver badges252 bronze badges

Comments

fujiu · Accepted Answer · 2019-02-24 15:49:31Z

2

Using a for-loop is indeed not the best option here, because it tends to be both slow and hard to read.

This kind of task is perfect for the dplyr package using pipes:

library(dplyr)

df_result <- d %>%
  group_by(State) %>%
  summarise(y = sum(Num),
            drilldown = tolower(first(State)))

print(df_result)

gives:

# A tibble: 2 x 3
  State       y drilldown
  <fct>   <dbl> <chr>    
1 ALABAMA    11 alabama  
2 ALASKA     15 alaska

If you want to achieve a nested list like in your example output instead of a data.frame, you can additionally use transpose() from the purrr package:

library(purrr)

transpose(df_result)

gives:

[[1]]
[[1]]$State
[1] "ALABAMA"

[[1]]$y
[1] 11

[[1]]$drilldown
[1] "alabama"


[[2]]
[[2]]$State
[1] "ALASKA"

[[2]]$y
[1] 15

[[2]]$drilldown
[1] "alaska"

edited Feb 24, 2019 at 15:49

answered Feb 24, 2019 at 15:00

fujiu

5014 silver badges9 bronze badges

3 Comments

Darren Tsai Over a year ago

Nice work! But the expected output should be a list, not a data frame. Maybe you can add some commands to transform it.

firmo23 Over a year ago

I made an edit. Yous solution gives 1 list with 2 'list of 3' while I would like 2 lists.

fujiu Over a year ago

Sorry, not sure if I understand. As far as I can tell, this leads to the exact same result as what you gave as your 'manual solution'. Could you be more specific?

Rui Barradas · Accepted Answer · 2019-02-24 16:41:40Z

2

Do you want something like the following? It uses split to make a list of dataframes, one per State and then lapply an anonymous function to each of the df's.

Here are two versions, with different output formats, an object of class "list" an an object of class "data.frame".

lapply(split(d, d$State), function(DF){
  s <- as.character(DF[["State"]][1])
  list(
    State = s,
    y = sum(DF[["Num"]]),
    drilldown = tolower(s)
  )
})

lapply(split(d, d$State), function(DF){
  s <- as.character(DF[["State"]][1])
  data.frame(
    State = s,
    y = sum(DF[["Num"]]),
    drilldown = tolower(s)
  )
})

EDIT.

The first way outputs a list of two named lists. If you want those lists to be unnamed, do what G. Grothendiek comment proposes.

data <-lapply(unname(split(d, d$State)), function(DF){
  s <- as.character(DF[["State"]][1])
  list(
    State = s,
    y = sum(DF[["Num"]]),
    drilldown = tolower(s)
  )
})

edited Feb 24, 2019 at 16:41

answered Feb 24, 2019 at 14:56

Rui Barradas

77.9k8 gold badges41 silver badges75 bronze badges

3 Comments

firmo23 Over a year ago

I made an edit. Yous solution gives 1 list with 2 'list of 3' while I would like 2 lists.

G. Grothendieck Over a year ago

This could be written: lapply(unname(split(d, d$State)), with, list(name = State[1], y = tolower(State[1]), drilldown = sum(Num)))

Rui Barradas Over a year ago

@firmo23 My first solution does output a list of two lists. Anyway, I have edited, see if this is it.

Collectives™ on Stack Overflow

Use loop to create lists inside another list

4 Answers 4

2 Comments

Comments

3 Comments

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

2 Comments

Comments

3 Comments

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related