storing output of for loop to a list in R

Question

I'm having trouble outputting the results of a for loop to a list/vector in R. The loop is running over a df structured as below, where each unique ID is represented by 1 to n rows:

id <- c(1, 2, 2, 2, 3, 4, 5, 6)
string <-c("apple", "grape", "orange", "blueberry", "plum", "tomato", "pear", "plum")
df <- data.frame(id, string)

For each unique ID, I want to write a list collapsing the n rows into a single row containing a concatenated character string based on the information in column "string". So I have:

#write a function to concatenate strings where d = dataframe, n = column name, and s = character to act as separator
concat <- function(d, n, s) {
   list_value = paste0(d[[n]], sep = s)
   return(list_value)
}

#create two empty lists
string_list <- list()
item_list <- list()

#loop the concatenate function over each unique id in the df
for (i in unique(df$id)) {

   item <- filter(df, id == i)
   print(item)
   item_list[i] <- item

   strings <- concat(item, "string", ";") 
   print(strings)
   string_list[i] <- strings

   }

I can see from the print statements that the loop is running "correctly" (I'm getting the output I want printed to the console) but I get warnings that "number of items to replace is not a multiple of replacement length" and string_list and item_list are impossibly large objects (a df of ~2000 rows becomes a list of ~10M elements).

If at the beginning of the loop I instead say:

 for (i in 1:length(df$id))

I get a list that is the same length as the number of rows in the original df; but it's empty (it returns integer [0] or character [1] for all). There are no NAs in the original df (checked with table(is.na(df$col_name)) for all columns). Same warnings.

Using string_list <- c() instead of string_list <- list() does not seem to help.

I'm missing something simple. What is it? Thanks

EDIT: I think I see part of the problem. The object "item" is a (small) df, and appending a series of dfs to a list would result in a large object. But replacing item_list <- list() with

item_data <- data.frame(Col1 = integer(), Col2 = character(), stringsAsFactors = FALSE)

gives an error, new columns would leave holes after existing columns

Try converting your item_list[ and string_list[ assignments to [[. — Chris
– Chris, Commented Aug 31, 2021 at 12:56
Do you mean item_list[[i] <- items and string_list[[i]] <- strings inside the loop? With for (i in unique(df$id)) {, that gives the same size 10M element lists. Interestingly, with for (i in 1:length(df$id)), the output now becomes large; two columns - an empty list of 0x3 and a df of 0x3 — user237554
– user237554, Commented Aug 31, 2021 at 13:19
@denisafonin the goal is for string_list to be ("apple;", "grape;" "orange;" "blueberry;", "plum;", "tomato;", "pear;", "plum;") and item_list to be (1, 2, 3, 4, 5, 6) so that they can be rejoined df2 <- data.frame(item_list, string_list), collapsing the n rows into a single row. I can't just drop duplicates on the original df because I need the information in the other rows. — user237554
– user237554, Commented Aug 31, 2021 at 13:24

denisafonin · Accepted Answer · 2021-08-31 13:43:37Z

1

Would this achieve the result you are looking for?

library(dplyr)

id <- c(1, 2, 2, 2, 3, 4, 5, 6)
string <-c("apple", "grape", "orange", "blueberry", "plum", "tomato", "pear", "plum")
df <- data.frame(id, string)

df2 <- df %>%
  group_by(id) %>%
  summarize(string = paste0(string, collapse = '; '))

Output:

> df2
# A tibble: 6 x 2
     id string                  
  <dbl> <chr>                   
1     1 apple                   
2     2 grape; orange; blueberry
3     3 plum                    
4     4 tomato                  
5     5 pear                    
6     6 plum

answered Aug 31, 2021 at 13:43

denisafonin

1,1361 gold badge7 silver badges17 bronze badges

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

Collectives™ on Stack Overflow

storing output of for loop to a list in R

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related