r - cbind() or merge() my data in 2 separate data frames whilst in a for loop - Stack Overflow

admin管理员组

文章数量:1402775

I wish to add some columns to my original data frame whilst looping through the rows in a loop. My issue is that when looping through I cant save out my dataframe after the cbind() my position as the rest of my dataset has an unequal number of columns, i tried to combat this by creating the column names first and NA-ing them before looping through my data but still no joy. Some examples of what ive tried at the bottom. Note...I know the following example doesn't need a loop its just a subset of a bigger thing I'm working with

myDat
id   name    city
1    john    london
2    james   manchester
3    steve   hull
4    david   birmingham

myDat2
att1     att2
23       21
25       30
44       28
30       28


for(i in 1:nrow(myDat)){
 #grab data that creates myDat 2 here then combine them
 #in this instance myDat2 is just line 1
 myDat2 = data.frame(att1 = 23, att2 = 21) #when i=1
 myDat[i,] = cbind(myDat[i, ], mydat2)
}

error/warning message = number of items to replace is not a multiple of replacement length

I wish to add some columns to my original data frame whilst looping through the rows in a loop. My issue is that when looping through I cant save out my dataframe after the cbind() my position as the rest of my dataset has an unequal number of columns, i tried to combat this by creating the column names first and NA-ing them before looping through my data but still no joy. Some examples of what ive tried at the bottom. Note...I know the following example doesn't need a loop its just a subset of a bigger thing I'm working with

myDat
id   name    city
1    john    london
2    james   manchester
3    steve   hull
4    david   birmingham

myDat2
att1     att2
23       21
25       30
44       28
30       28


for(i in 1:nrow(myDat)){
 #grab data that creates myDat 2 here then combine them
 #in this instance myDat2 is just line 1
 myDat2 = data.frame(att1 = 23, att2 = 21) #when i=1
 myDat[i,] = cbind(myDat[i, ], mydat2)
}

error/warning message = number of items to replace is not a multiple of replacement length

• r

Share Improve this question Follow asked Mar 21 at 14:12 JoeJoe 1,39755 silver badges2121 bronze badges

Add a comment  | 

2 Answers 2

Sorted by: Reset to default 2

You can first add a list column and store pretty much any object at any index there. If that object happens to be a frame (or frame-like list, for example), you can later unnest it to a regular flat data.frame:

myDat$list_col <- vector(mode = "list", length = nrow(myDat))

for(i in 1:nrow(myDat)){
  myDat2 = data.frame(att1 = 23, att2 = 21)
  if (i %% 2 == 0) myDat2 = cbind(myDat2, att3 = 33)
  myDat$list_col[[i]] = myDat2
}

myDat
#>   id  name       city   list_col
#> 1  1  john     london     23, 21
#> 2  2 james manchester 23, 21, 33
#> 3  3 steve       hull     23, 21
#> 4  4 david birmingham 23, 21, 33

tidyr::unnest_wider(myDat, list_col)
#> # A tibble: 4 × 6
#>      id name  city        att1  att2  att3
#>   <int> <chr> <chr>      <dbl> <dbl> <dbl>
#> 1     1 john  london        23    21    NA
#> 2     2 james manchester    23    21    33
#> 3     3 steve hull          23    21    NA
#> 4     4 david birmingham    23    21    33

---

Example data:

myDat <- read.table(header = TRUE, text = 
"id   name    city
1    john    london
2    james   manchester
3    steve   hull
4    david   birmingham")

myDat2 <- read.table(header = TRUE, text = 
"att1     att2
23       21
25       30
44       28
30       28")

Another option might be to use data.table, which allows assigning a data.frame/data.table to columns in another table. For example:

library(data.table)
myDat = data.table(id = 1:4, 
                   name=c("john", "james", "steve", "david"),    
                   city=c("london", "manchester", "hull", "birmingham")) 

for(i in 1:nrow(myDat)){
    myDat2 <- data.frame(att1 = c(23, 25, 44, 30)[i], att2 = c(21, 30, 28, 28)[i])
    myDat[i, c("att1", "att2"):=myDat2]
}
print(myDat)

Result

#>       id   name       city  att1  att2
#>    <int> <char>     <char> <num> <num>
#> 1:     1   john     london    23    21
#> 2:     2  james manchester    25    30
#> 3:     3  steve       hull    44    28
#> 4:     4  david birmingham    30    28

本文标签： rcbind() or merge() my data in 2 separate data frames whilst in a for loopStack Overflow