Быстрый способ сопоставить фрейм данных с низким разрешением с фреймом данных с более высоким разрешением по столбцам даты и времени в R - PullRequest
Новая
       54

Быстрый способ сопоставить фрейм данных с низким разрешением с фреймом данных с более высоким разрешением по столбцам даты и времени в R

0 голосов
/ 26 сентября 2019

Я пытаюсь сопоставить строки в одном кадре данных (df1) с строками в другом (df2) на основе столбца datetime.Мой желаемый вывод - это кадр данных df1 с соответствующими строками из df2, добавленными к нему на основе временных меток.Интервал времени данных в df1 имеет более высокое разрешение, чем в df2.У меня есть некоторый код, который находит ближайшую запись в df2, до или после записи в df1, чтобы делать то, что я хочу, однако количество строк в каждом из моих файлов df1 велико (> 500 000), и у меня есть 100 пар файловчтобы соответствовать.У меня слишком длинный цикл for.До сих пор он работал на одной и той же паре файлов в течение 24 часов!Я знаю, что это работает, потому что я взял небольшие подмножества данных и быстро сопоставил их, и я использовал это для цикла несколько раз на разных наборах данных.Я задаюсь вопросом, узнает ли кто-нибудь о более быстром и более элегантном способе сделать это, поскольку с такой скоростью я никогда не объединю файлы!Любые предложения будут ценны.Большое спасибо.

Текущий код, который работает вечно: 

# The function that will merge the two dataframes
addSuppInfo = function(dfs1, dfs2){
  N = dim(dfs1)[1]
  MergedDF = as.data.frame(array(NA,c(N,length(names(dfs1))+length(names(dfs2)))))
  names(MergedDF) = c(names(dfs1),names(dfs2))
  suppdate = which(names(MergedDF)=="datetime")[2]  #The new, empty database has two "datetime" columns now, lets rename the second to avoid confusion
  names(MergedDF)[suppdate] = "GPS_datetime_local"

  #Convert datetimes to some universal measure of time so there can be no confusion
  dfs1$AbsTime = unclass(as.POSIXct(strptime(dfs1$datetime, "%Y-%m-%d %H:%M:%S"))) 
  dfs2$AbsTime = unclass(as.POSIXct(strptime(dfs2$datetime_utc, "%Y-%m-%d %H:%M:%S "))) 
  for(record in 1:N){
    #Find the closest record, whether before or after a time in the supp info
    NearestRow = which.min(abs(dfs2$AbsTime - dfs1$AbsTime[record]))[1] 
    if(length(NearestRow) > 0){ 
      NewRow = cbind(dfs1[record,], dfs2[NearestRow,])
      #We have two AbsTime columns which we're just using for the lookup so remove them
      NewRow = NewRow[which(names(NewRow) != "AbsTime")]  
      MergedDF[record,] = NewRow
    }else{  #if there is no matching supp data in df2, just put the df1 data in the row with a "no data" label. Bit fiddly, but neater than just leaving rows blank
      MergedDF[record,(1:(length(names(dfs1))-1))] = dfs1[record, (1:(length(names(dfs1))-1))]  #minus one because we don't want the AbsTime column we added on earlier
      MergedDF[record,length(names(dfs1)):length(names(MergedDF))] = "NA"
    }
  }
  return(MergedDF)  
}



# Create lists of the df1 and df2 filenames (for my case where I ahve lots of files in a folder).
dir = "/Users/rhiannonaustin/Desktop/Booby_analysis/EMbC/10sec_Interp_Data/Merged/Match_TDR_IMM_GPS/BB/Raw GPS match/2017/GPS_Imm/test" 
setwd(dir) 
df2files = list.files(path=dir, pattern="*_df2.csv", full.names=FALSE, recursive=FALSE)
df1files = list.files(path=dir, pattern="*_df1.csv", full.names=FALSE, recursive=FALSE)


# For each file in those lists, submit them to our addSuppInfo function
# end up with df1 with new columns for the closest matching df2 row
for(i in 1:length(df2files)){
  file = df2files[i]
  #Extract the code
  df2code = strsplit(file,"_")[[1]][1]
  #Double check there is a matching df2 file
  df1code = strsplit(df1files[i],"_")[[1]][1]
  if(df2code == df1code){
    df1Data = read.csv(df1files[i], header=TRUE, stringsAsFactors = FALSE)
    df2Data = read.csv(file, header=TRUE, stringsAsFactors = FALSE)
    Merged = addSuppInfo(df1Data, df2Data)
    outfile = paste(df2code,"_merged.csv",sep="")
    write.csv(Merged, file=outfile, row.names=FALSE)
  }else{
    print(paste("No matching data for ",df2code,"!!"))
  }      
}

Пример данных (очевидно, этот небольшой пример работает быстро с моим собственным циклом for, но в реальной жизни мои наборы данных НАМНОГО больше!): 

    > dput(df1Data)
    structure(list(iid = 14554:14677, datetime = c("2017-03-12 15:55:04", 
    "2017-03-12 15:55:05", "2017-03-12 15:55:06", "2017-03-12 15:55:07", 
    "2017-03-12 15:55:08", "2017-03-12 15:55:09", "2017-03-12 15:55:10", 
    "2017-03-12 15:55:11", "2017-03-12 15:55:12", "2017-03-12 15:55:13", 
    "2017-03-12 15:55:14", "2017-03-12 15:55:15", "2017-03-12 15:55:16", 
    "2017-03-12 15:55:17", "2017-03-12 15:55:18", "2017-03-12 15:55:19", 
    "2017-03-12 15:55:20", "2017-03-12 15:55:21", "2017-03-12 15:55:22", 
    "2017-03-12 15:55:23", "2017-03-12 15:55:24", "2017-03-12 15:55:25", 
    "2017-03-12 15:55:26", "2017-03-12 15:55:27", "2017-03-12 15:55:28", 
    "2017-03-12 15:55:29", "2017-03-12 15:55:30", "2017-03-12 15:55:31", 
    "2017-03-12 15:55:32", "2017-03-12 15:55:33", "2017-03-12 15:55:34", 
    "2017-03-12 15:55:35", "2017-03-12 15:55:36", "2017-03-12 15:55:37", 
    "2017-03-12 15:55:38", "2017-03-12 15:55:39", "2017-03-12 15:55:40", 
    "2017-03-12 15:55:41", "2017-03-12 15:55:42", "2017-03-12 15:55:43", 
    "2017-03-12 15:55:44", "2017-03-12 15:55:45", "2017-03-12 15:55:46", 
    "2017-03-12 15:55:47", "2017-03-12 15:55:48", "2017-03-12 15:55:49", 
    "2017-03-12 15:55:50", "2017-03-12 15:55:51", "2017-03-12 15:55:52", 
    "2017-03-12 15:55:53", "2017-03-12 15:55:54", "2017-03-12 15:55:55", 
    "2017-03-12 15:55:56", "2017-03-12 15:55:57", "2017-03-12 15:55:58", 
    "2017-03-12 15:55:59", "2017-03-12 15:56:00", "2017-03-12 15:56:01", 
    "2017-03-12 15:56:02", "2017-03-12 15:56:03", "2017-03-12 15:56:04", 
    "2017-03-12 15:56:05", "2017-03-12 15:56:06", "2017-03-12 15:56:07", 
    "2017-03-12 15:56:08", "2017-03-12 15:56:09", "2017-03-12 15:56:10", 
    "2017-03-12 15:56:11", "2017-03-12 15:56:12", "2017-03-12 15:56:13", 
    "2017-03-12 15:56:14", "2017-03-12 15:56:15", "2017-03-12 15:56:16", 
    "2017-03-12 15:56:17", "2017-03-12 15:56:18", "2017-03-12 15:56:19", 
    "2017-03-12 15:56:20", "2017-03-12 15:56:21", "2017-03-12 15:56:22", 
    "2017-03-12 15:56:23", "2017-03-12 15:56:24", "2017-03-12 15:56:25", 
    "2017-03-12 15:56:26", "2017-03-12 15:56:27", "2017-03-12 15:56:28", 
    "2017-03-12 15:56:29", "2017-03-12 15:56:30", "2017-03-12 15:56:31", 
    "2017-03-12 15:56:32", "2017-03-12 15:56:33", "2017-03-12 15:56:34", 
    "2017-03-12 15:56:35", "2017-03-12 15:56:36", "2017-03-12 15:56:37", 
    "2017-03-12 15:56:38", "2017-03-12 15:56:39", "2017-03-12 15:56:40", 
    "2017-03-12 15:56:41", "2017-03-12 15:56:42", "2017-03-12 15:56:43", 
    "2017-03-12 15:56:44", "2017-03-12 15:56:45", "2017-03-12 15:56:46", 
    "2017-03-12 15:56:47", "2017-03-12 15:56:48", "2017-03-12 15:56:49", 
    "2017-03-12 15:56:50", "2017-03-12 15:56:51", "2017-03-12 15:56:52", 
    "2017-03-12 15:56:53", "2017-03-12 15:56:54", "2017-03-12 15:56:55", 
    "2017-03-12 15:56:56", "2017-03-12 15:56:57", "2017-03-12 15:56:58", 
    "2017-03-12 15:56:59", "2017-03-12 15:57:00", "2017-03-12 15:57:01", 
    "2017-03-12 15:57:02", "2017-03-12 15:57:03", "2017-03-12 15:57:04", 
    "2017-03-12 15:57:05", "2017-03-12 15:57:06", "2017-03-12 15:57:07"
    ), state = c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 
    2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L
    )), class = "data.frame", row.names = c(NA, -124L))

> dput(df2Data)
structure(list(new_id = 188889:188903, datetime_utc = c("2017-03-12 15:52:58", 
"2017-03-12 15:53:28", "2017-03-12 15:53:58", "2017-03-12 15:54:28", 
"2017-03-12 15:54:58", "2017-03-12 15:55:28", "2017-03-12 15:55:58", 
"2017-03-12 15:56:28", "2017-03-12 15:56:58", "2017-03-12 15:57:28", 
"2017-03-12 15:57:58", "2017-03-12 15:58:28", "2017-03-12 15:58:58", 
"2017-03-12 15:59:28", "2017-03-12 15:59:58"), count = c(6.138159, 
0.6693379, 5.507328, 4.041119, 1.464152, 11.73605, 12.92626, 
1.397087, 3.032117, 1.913028, 2.029836, 9.386913, 7.672296, 6.36146, 
0.5731061)), class = "data.frame", row.names = c(NA, -15L))

Желаемый вывод: 

> Merged
      iid            datetime state new_id        datetime_utc     count
1   14554 2017-03-12 15:55:04     1 188893 2017-03-12 15:54:58  1.464152
2   14555 2017-03-12 15:55:05     1 188893 2017-03-12 15:54:58  1.464152
3   14556 2017-03-12 15:55:06     1 188893 2017-03-12 15:54:58  1.464152
4   14557 2017-03-12 15:55:07     1 188893 2017-03-12 15:54:58  1.464152
5   14558 2017-03-12 15:55:08     1 188893 2017-03-12 15:54:58  1.464152
6   14559 2017-03-12 15:55:09     1 188893 2017-03-12 15:54:58  1.464152
7   14560 2017-03-12 15:55:10     1 188893 2017-03-12 15:54:58  1.464152
8   14561 2017-03-12 15:55:11     1 188893 2017-03-12 15:54:58  1.464152
9   14562 2017-03-12 15:55:12     1 188893 2017-03-12 15:54:58  1.464152
10  14563 2017-03-12 15:55:13     1 188893 2017-03-12 15:54:58  1.464152
11  14564 2017-03-12 15:55:14     1 188894 2017-03-12 15:55:28 11.736050
12  14565 2017-03-12 15:55:15     1 188894 2017-03-12 15:55:28 11.736050
13  14566 2017-03-12 15:55:16     1 188894 2017-03-12 15:55:28 11.736050
14  14567 2017-03-12 15:55:17     1 188894 2017-03-12 15:55:28 11.736050
15  14568 2017-03-12 15:55:18     1 188894 2017-03-12 15:55:28 11.736050
16  14569 2017-03-12 15:55:19     1 188894 2017-03-12 15:55:28 11.736050
17  14570 2017-03-12 15:55:20     1 188894 2017-03-12 15:55:28 11.736050
18  14571 2017-03-12 15:55:21     1 188894 2017-03-12 15:55:28 11.736050
19  14572 2017-03-12 15:55:22     1 188894 2017-03-12 15:55:28 11.736050
20  14573 2017-03-12 15:55:23     1 188894 2017-03-12 15:55:28 11.736050
21  14574 2017-03-12 15:55:24     1 188894 2017-03-12 15:55:28 11.736050
22  14575 2017-03-12 15:55:25     1 188894 2017-03-12 15:55:28 11.736050
23  14576 2017-03-12 15:55:26     1 188894 2017-03-12 15:55:28 11.736050
24  14577 2017-03-12 15:55:27     1 188894 2017-03-12 15:55:28 11.736050
25  14578 2017-03-12 15:55:28     1 188894 2017-03-12 15:55:28 11.736050
26  14579 2017-03-12 15:55:29     1 188894 2017-03-12 15:55:28 11.736050
27  14580 2017-03-12 15:55:30     1 188894 2017-03-12 15:55:28 11.736050
28  14581 2017-03-12 15:55:31     1 188894 2017-03-12 15:55:28 11.736050
29  14582 2017-03-12 15:55:32     1 188894 2017-03-12 15:55:28 11.736050
30  14583 2017-03-12 15:55:33     1 188894 2017-03-12 15:55:28 11.736050
31  14584 2017-03-12 15:55:34     1 188894 2017-03-12 15:55:28 11.736050
32  14585 2017-03-12 15:55:35     1 188894 2017-03-12 15:55:28 11.736050
33  14586 2017-03-12 15:55:36     1 188894 2017-03-12 15:55:28 11.736050
34  14587 2017-03-12 15:55:37     1 188894 2017-03-12 15:55:28 11.736050
35  14588 2017-03-12 15:55:38     1 188894 2017-03-12 15:55:28 11.736050
36  14589 2017-03-12 15:55:39     1 188894 2017-03-12 15:55:28 11.736050
37  14590 2017-03-12 15:55:40     1 188894 2017-03-12 15:55:28 11.736050
38  14591 2017-03-12 15:55:41     1 188894 2017-03-12 15:55:28 11.736050
39  14592 2017-03-12 15:55:42     1 188894 2017-03-12 15:55:28 11.736050
40  14593 2017-03-12 15:55:43     1 188894 2017-03-12 15:55:28 11.736050
41  14594 2017-03-12 15:55:44     1 188895 2017-03-12 15:55:58 12.926260
42  14595 2017-03-12 15:55:45     1 188895 2017-03-12 15:55:58 12.926260
43  14596 2017-03-12 15:55:46     1 188895 2017-03-12 15:55:58 12.926260
44  14597 2017-03-12 15:55:47     1 188895 2017-03-12 15:55:58 12.926260
45  14598 2017-03-12 15:55:48     1 188895 2017-03-12 15:55:58 12.926260
46  14599 2017-03-12 15:55:49     1 188895 2017-03-12 15:55:58 12.926260
47  14600 2017-03-12 15:55:50     1 188895 2017-03-12 15:55:58 12.926260
48  14601 2017-03-12 15:55:51     1 188895 2017-03-12 15:55:58 12.926260
49  14602 2017-03-12 15:55:52     1 188895 2017-03-12 15:55:58 12.926260
50  14603 2017-03-12 15:55:53     1 188895 2017-03-12 15:55:58 12.926260
51  14604 2017-03-12 15:55:54     1 188895 2017-03-12 15:55:58 12.926260
52  14605 2017-03-12 15:55:55     1 188895 2017-03-12 15:55:58 12.926260
53  14606 2017-03-12 15:55:56     1 188895 2017-03-12 15:55:58 12.926260
54  14607 2017-03-12 15:55:57     1 188895 2017-03-12 15:55:58 12.926260
55  14608 2017-03-12 15:55:58     1 188895 2017-03-12 15:55:58 12.926260
56  14609 2017-03-12 15:55:59     1 188895 2017-03-12 15:55:58 12.926260
57  14610 2017-03-12 15:56:00     1 188895 2017-03-12 15:55:58 12.926260
58  14611 2017-03-12 15:56:01     1 188895 2017-03-12 15:55:58 12.926260
59  14612 2017-03-12 15:56:02     1 188895 2017-03-12 15:55:58 12.926260
60  14613 2017-03-12 15:56:03     1 188895 2017-03-12 15:55:58 12.926260
61  14614 2017-03-12 15:56:04     1 188895 2017-03-12 15:55:58 12.926260
62  14615 2017-03-12 15:56:05     1 188895 2017-03-12 15:55:58 12.926260
63  14616 2017-03-12 15:56:06     1 188895 2017-03-12 15:55:58 12.926260
64  14617 2017-03-12 15:56:07     1 188895 2017-03-12 15:55:58 12.926260
65  14618 2017-03-12 15:56:08     1 188895 2017-03-12 15:55:58 12.926260
66  14619 2017-03-12 15:56:09     1 188895 2017-03-12 15:55:58 12.926260
67  14620 2017-03-12 15:56:10     1 188895 2017-03-12 15:55:58 12.926260
68  14621 2017-03-12 15:56:11     1 188895 2017-03-12 15:55:58 12.926260
69  14622 2017-03-12 15:56:12     1 188895 2017-03-12 15:55:58 12.926260
70  14623 2017-03-12 15:56:13     1 188895 2017-03-12 15:55:58 12.926260
71  14624 2017-03-12 15:56:14     1 188896 2017-03-12 15:56:28  1.397087
72  14625 2017-03-12 15:56:15     1 188896 2017-03-12 15:56:28  1.397087
73  14626 2017-03-12 15:56:16     1 188896 2017-03-12 15:56:28  1.397087
74  14627 2017-03-12 15:56:17     1 188896 2017-03-12 15:56:28  1.397087
75  14628 2017-03-12 15:56:18     1 188896 2017-03-12 15:56:28  1.397087
76  14629 2017-03-12 15:56:19     1 188896 2017-03-12 15:56:28  1.397087
77  14630 2017-03-12 15:56:20     1 188896 2017-03-12 15:56:28  1.397087
78  14631 2017-03-12 15:56:21     1 188896 2017-03-12 15:56:28  1.397087
79  14632 2017-03-12 15:56:22     1 188896 2017-03-12 15:56:28  1.397087
80  14633 2017-03-12 15:56:23     1 188896 2017-03-12 15:56:28  1.397087
81  14634 2017-03-12 15:56:24     1 188896 2017-03-12 15:56:28  1.397087
82  14635 2017-03-12 15:56:25     1 188896 2017-03-12 15:56:28  1.397087
83  14636 2017-03-12 15:56:26     1 188896 2017-03-12 15:56:28  1.397087
84  14637 2017-03-12 15:56:27     1 188896 2017-03-12 15:56:28  1.397087
85  14638 2017-03-12 15:56:28     1 188896 2017-03-12 15:56:28  1.397087
86  14639 2017-03-12 15:56:29     1 188896 2017-03-12 15:56:28  1.397087
87  14640 2017-03-12 15:56:30     1 188896 2017-03-12 15:56:28  1.397087
88  14641 2017-03-12 15:56:31     1 188896 2017-03-12 15:56:28  1.397087
89  14642 2017-03-12 15:56:32     1 188896 2017-03-12 15:56:28  1.397087
90  14643 2017-03-12 15:56:33     1 188896 2017-03-12 15:56:28  1.397087
91  14644 2017-03-12 15:56:34     1 188896 2017-03-12 15:56:28  1.397087
92  14645 2017-03-12 15:56:35     1 188896 2017-03-12 15:56:28  1.397087
93  14646 2017-03-12 15:56:36     1 188896 2017-03-12 15:56:28  1.397087
94  14647 2017-03-12 15:56:37     1 188896 2017-03-12 15:56:28  1.397087
95  14648 2017-03-12 15:56:38     1 188896 2017-03-12 15:56:28  1.397087
96  14649 2017-03-12 15:56:39     2 188896 2017-03-12 15:56:28  1.397087
97  14650 2017-03-12 15:56:40     2 188896 2017-03-12 15:56:28  1.397087
98  14651 2017-03-12 15:56:41     2 188896 2017-03-12 15:56:28  1.397087
99  14652 2017-03-12 15:56:42     2 188896 2017-03-12 15:56:28  1.397087
100 14653 2017-03-12 15:56:43     2 188896 2017-03-12 15:56:28  1.397087
101 14654 2017-03-12 15:56:44     2 188897 2017-03-12 15:56:58  3.032117
102 14655 2017-03-12 15:56:45     2 188897 2017-03-12 15:56:58  3.032117
103 14656 2017-03-12 15:56:46     2 188897 2017-03-12 15:56:58  3.032117
104 14657 2017-03-12 15:56:47     2 188897 2017-03-12 15:56:58  3.032117
105 14658 2017-03-12 15:56:48     2 188897 2017-03-12 15:56:58  3.032117
106 14659 2017-03-12 15:56:49     2 188897 2017-03-12 15:56:58  3.032117
107 14660 2017-03-12 15:56:50     2 188897 2017-03-12 15:56:58  3.032117
108 14661 2017-03-12 15:56:51     2 188897 2017-03-12 15:56:58  3.032117
109 14662 2017-03-12 15:56:52     2 188897 2017-03-12 15:56:58  3.032117
110 14663 2017-03-12 15:56:53     2 188897 2017-03-12 15:56:58  3.032117
111 14664 2017-03-12 15:56:54     2 188897 2017-03-12 15:56:58  3.032117
112 14665 2017-03-12 15:56:55     2 188897 2017-03-12 15:56:58  3.032117
113 14666 2017-03-12 15:56:56     2 188897 2017-03-12 15:56:58  3.032117
114 14667 2017-03-12 15:56:57     2 188897 2017-03-12 15:56:58  3.032117
115 14668 2017-03-12 15:56:58     2 188897 2017-03-12 15:56:58  3.032117
116 14669 2017-03-12 15:56:59     2 188897 2017-03-12 15:56:58  3.032117
117 14670 2017-03-12 15:57:00     2 188897 2017-03-12 15:56:58  3.032117
118 14671 2017-03-12 15:57:01     2 188897 2017-03-12 15:56:58  3.032117
119 14672 2017-03-12 15:57:02     2 188897 2017-03-12 15:56:58  3.032117
120 14673 2017-03-12 15:57:03     2 188897 2017-03-12 15:56:58  3.032117
121 14674 2017-03-12 15:57:04     2 188897 2017-03-12 15:56:58  3.032117
122 14675 2017-03-12 15:57:05     2 188897 2017-03-12 15:56:58  3.032117
123 14676 2017-03-12 15:57:06     2 188897 2017-03-12 15:56:58  3.032117
124 14677 2017-03-12 15:57:07     2 188897 2017-03-12 15:56:58  3.032117

1 Ответ

0 голосов
/ 27 сентября 2019

Я немного поиграл, рабочий код с вашими данными выглядел бы так: 

library(tidyverse)
library(rlang)

df1Data <- df1Data %>%
  as_tibble() %>%
  mutate(datetime = as.POSIXct(datetime))

df2Data <- df2Data %>%
  as_tibble() %>%
  mutate(datetime_utc = as.POSIXct(datetime_utc))

# Loop through every time entry from df1Data and find nearest time stamp in df2Data.
# Insert "new_id" from df2Data in df1Data as grouping column.
df1Data <- mutate(df1Data, df2Row = sapply(1:nrow(df1Data), function(i) 
  df2Data$new_id[which.min(abs(.data$datetime[i] - df2Data$datetime_utc))])
  )

# Merge data frames by grouping column "df2Row" from df1Data und "new_id" from df2Data
mergedData <- left_join(df1Data, df2Data, by = c("df2Row" = "new_id")) %>%
  rename("new_id" = "df2Row")

Я не знаю производительность вашего большого набора данных, вы могли бы попробовать.Если это будет слишком медленно, мы могли бы двигаться вперед с rcpp.

...