堆栈内存溢出
  简体   繁体   English

后续行两列之间的因子变量的时差

[英]Time Difference for a factor variable between two columns on subsequent rows

 原文  2017-08-18 10:33:44  7   4    r/ datetime/ data.table

问题描述

I have a data frame with Entry Time and Exit time . 我有一个包含Entry Time和Exit time的数据框。 I am trying to get a column say Stay Duration that is difference between exit time and next row Entry Time. 我正在尝试获得一列Stay Duration ，该Stay Duration是退出时间与下一行进入时间之间的差。 dataframe is sorted by Date and Entry Time. 数据框按日期和输入时间排序。

For the first row Stay Duration is: 6:54:50 - 7:34:36 = 0:39:46 对于第一行，停留时间为：6: 6:54:50 - 7:34:36 = 0:39:46 : 6:54:50 - 7:34:36 = 0:39:46 : 6:54:50 - 7:34:36 = 0:39:46

The problem i face is for the below scenario: when Exit_Time is 22:15:05 and next row Entry_Time has 6:02:46 . 我遇到的问题是针对以下情况：当Exit_Time is 22:15:05且下一行Entry_Time has 6:02:46 。 Fr this case i need the Stay Duration to be 7:47:41 如果是这种情况，我需要Stay Duration to be 7:47:41 

dput(df)
structure(list(JRNY_ID_NUM = c(115492027250, 115492027250, 115523231209, 
115523231209, 115526742250, 115526742250, 115509240124, 115509240124, 
115539253765, 115539253765, 115570245886, 115567046025, 115562452408, 
115562452408, 115574565032, 115574565032), BIZ_DT = structure(c(1L, 
1L, 1L, 1L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 4L, 4L, 4L, 4L), .Label = c("2017-01-01", 
"2017-01-02", "2017-01-03", "2017-01-04", "2017-01-05", "2017-01-06", 
"2017-01-07", "2017-01-09", "2017-01-10", "2017-01-11", "2017-01-12", 
"2017-01-13", "2017-01-14", "2017-01-15", "2017-01-16", "2017-01-17", 
"2017-01-18", "2017-01-19", "2017-01-20", "2017-01-21", "2017-01-22", 
"2017-01-23", "2017-01-24", "2017-01-25", "2017-01-26", "2017-01-27", 
"2017-01-31"), class = "factor"), ENTRY_TM = structure(c(37L, 
41L, 45L, 46L, 8L, 25L, 52L, 73L, 5L, 15L, 56L, 89L, 29L, 33L, 
63L, 77L), .Label = c("05:30:39", "05:32:07", "05:32:33", "05:32:38", 
"05:32:50", "05:32:59", "05:33:06", "05:37:14", "05:37:58", "05:38:34", 
"05:38:38", "05:40:22", "05:40:49", "05:41:16", "05:42:27", "05:47:17", 
"05:48:03", "05:48:13", "05:48:54", "05:49:15", "05:50:17", "05:51:42", 
"05:52:30", "05:53:20", "05:54:40", "05:56:24", "05:57:59", "06:00:11", 
"06:02:46", "06:03:28", "06:05:44", "06:32:18", "06:40:32", "06:40:40", 
"06:42:35", "06:45:51", "06:45:55", "06:52:49", "06:57:25", "07:03:49", 
"07:34:36", "08:26:43", "09:16:34", "10:16:10", "12:21:51", "13:36:40", 
"15:29:30", "16:07:03", "16:10:49", "16:13:51", "16:15:04", "16:29:20", 
"16:47:49", "16:48:42", "16:55:50", "16:56:27", "16:58:53", "17:01:02", 
"17:03:31", "17:06:19", "17:09:03", "17:11:22", "17:12:15", "17:12:57", 
"17:15:11", "17:16:56", "17:21:07", "17:22:18", "17:22:22", "17:23:53", 
"17:28:37", "17:30:17", "17:30:24", "17:31:21", "17:32:22", "17:59:07", 
"18:16:25", "18:17:13", "18:23:36", "18:27:40", "18:44:43", "18:46:36", 
"18:53:21", "20:55:32", "21:06:00", "21:07:08", "21:18:10", "21:18:21", 
"21:42:25", "21:43:45", "23:31:38"), class = "factor"), EXIT_TM = structure(c(34L, 
37L, 45L, 46L, 9L, 27L, 54L, 60L, 7L, 20L, 71L, 88L, 25L, 40L, 
68L, 72L), .Label = c("?", "05:37:56", "05:39:50", "05:39:51", 
"05:39:53", "05:40:03", "05:40:51", "05:41:01", "05:43:02", "05:44:51", 
"05:45:04", "05:45:45", "05:56:35", "05:57:45", "05:58:58", "06:03:06", 
"06:21:39", "06:23:35", "06:24:30", "06:24:58", "06:28:36", "06:29:17", 
"06:29:25", "06:32:11", "06:32:15", "06:34:01", "06:36:28", "06:41:02", 
"06:41:05", "06:43:31", "06:44:44", "06:51:05", "06:51:46", "06:54:50", 
"07:07:55", "07:39:39", "07:43:21", "07:48:49", "08:05:22", "08:17:58", 
"08:18:15", "08:32:12", "09:34:23", "10:31:51", "13:17:38", "13:46:19", 
"16:08:29", "16:26:21", "16:26:50", "16:37:08", "17:09:13", "17:25:49", 
"17:26:31", "17:26:50", "17:27:35", "17:28:20", "17:31:16", "17:34:43", 
"17:35:16", "17:36:37", "17:42:09", "17:42:23", "18:00:27", "18:06:53", 
"18:08:38", "18:09:18", "18:13:02", "18:14:35", "18:20:15", "18:22:06", 
"18:23:17", "18:25:18", "18:25:30", "18:28:47", "18:30:11", "18:30:54", 
"18:33:31", "18:38:49", "18:41:19", "18:52:25", "19:05:37", "19:27:49", 
"21:06:21", "21:41:28", "21:47:13", "21:53:35", "21:54:29", "22:15:05", 
"22:25:41", "23:59:35"), class = "factor")), .Names = c("JRNY_ID_NUM", 
"BIZ_DT", "ENTRY_TM", "EXIT_TM"), row.names = c(160L, 73L, 51L, 
145L, 111L, 56L, 119L, 157L, 168L, 131L, 81L, 78L, 135L, 35L, 
165L, 25L), class = "data.frame")

The output i am trying to get is: 我想得到的输出是： 

JRNY_ID_NUM     BIZ_DT      ENTRY_TM    EXIT_TM     Stay_Duration
115492027250    1/1/2017    6:45:55     6:54:50     0:39:46
115492027250    1/1/2017    7:34:36     7:43:21     4:38:30
115523231209    1/1/2017    12:21:51    13:17:38    0:19:02
115523231209    1/1/2017    13:36:40    13:46:19    15:50:55
115526742250    1/2/2017    5:37:14     5:43:02     0:11:38
115526742250    1/2/2017    5:54:40     6:36:28     9:52:52
115509240124    1/2/2017    16:29:20    17:26:50    0:03:34
115509240124    1/2/2017    17:30:24    17:36:37    11:56:13
115539253765    1/3/2017    5:32:50     5:40:51     0:01:36
115539253765    1/3/2017    5:42:27     6:24:58     10:31:29
115570245886    1/3/2017    16:56:27    18:23:17    3:19:08
115567046025    1/3/2017    21:42:25    22:15:05    7:47:41
115562452408    1/4/2017    6:02:46     6:32:15     0:08:17
115562452408    1/4/2017    6:40:32     8:17:58     8:54:17
115574565032    1/4/2017    17:12:15    18:14:35    0:01:50
115574565032    1/4/2017    18:16:25    18:25:18

4 个解决方案

解决方案1
 1 已采纳  2017-08-18 11:06:24

Updated 更新

Try: 尝试： 

require(dplyr)

diff_to_hms <- function(x) {
    y <- abs(x)
    sprintf("%s:%02d:%02d:%02d", 
      ifelse(x < 0, "-", ""), 
      y %% 86400 %/% 3600,  
      y %% 3600 %/% 60,  
      y %% 60 %/% 1) 
}

dat %>% mutate(ENTRY_TM = as.POSIXct(strptime(paste(BIZ_DT,ENTRY_TM),format = "%Y-%m-%d %H:%M:%S")),
            EXIT_TM = as.POSIXct(strptime(paste(BIZ_DT,EXIT_TM),format = "%Y-%m-%d %H:%M:%S"))) %>%
            mutate(Stay_Duration = as.numeric(difftime(EXIT_TM,lead(ENTRY_TM),unit="secs")*-1),
                Stay_Duration2 = diff_to_hms(Stay_Duration))

解决方案2
 1  2017-08-18 11:11:53

try this: 尝试这个： 

require(data.table)
require(Hmisc)
setDT(df)
setnames(df, Cs(ID, da, en, ex))
df[, en1 := as.POSIXct(paste(da, en))]
df[, ex1 := as.POSIXct(paste(da, ex))]
df[, s := Lag(en1, -1)]
df[, Stay_Duration := s - ex1]
df[, Cs(en1,ex1,s) := NULL]
df
# ID         da       en       ex   Stay_Duration
# 1: 115492027250 2017-01-01 06:45:55 06:54:50  39.766667 mins
# 2: 115492027250 2017-01-01 07:34:36 07:43:21 278.500000 mins
# 3: 115523231209 2017-01-01 12:21:51 13:17:38  19.033333 mins
# 4: 115523231209 2017-01-01 13:36:40 13:46:19 950.916667 mins
# 5: 115526742250 2017-01-02 05:37:14 05:43:02  11.633333 mins
# 6: 115526742250 2017-01-02 05:54:40 06:36:28 592.866667 mins
# 7: 115509240124 2017-01-02 16:29:20 17:26:50   3.566667 mins
# 8: 115509240124 2017-01-02 17:30:24 17:36:37 716.216667 mins
# 9: 115539253765 2017-01-03 05:32:50 05:40:51   1.600000 mins
# 10: 115539253765 2017-01-03 05:42:27 06:24:58 631.483333 mins
# 11: 115570245886 2017-01-03 16:56:27 18:23:17 199.133333 mins
# 12: 115567046025 2017-01-03 21:42:25 22:15:05 467.683333 mins
# 13: 115562452408 2017-01-04 06:02:46 06:32:15   8.283333 mins
# 14: 115562452408 2017-01-04 06:40:32 08:17:58 534.283333 mins
# 15: 115574565032 2017-01-04 17:12:15 18:14:35   1.833333 mins
# 16: 115574565032 2017-01-04 18:16:25 18:25:18         NA mins

formatting: 格式： 

df[, stay2 := strftime(
  as.POSIXct(
    as.numeric(Stay_Duration)*60,
    origin = as.POSIXct("00:00:00", format = "%H:%M:%S", tz = "GMT"),
    tz = "GMT"),
  format = "%H:%M:%S", tz = "GMT")]
df
#              ID         da       en       ex   Stay_Duration    stay2
# 1: 115492027250 2017-01-01 06:45:55 06:54:50  39.766667 mins 00:39:46
# 2: 115492027250 2017-01-01 07:34:36 07:43:21 278.500000 mins 04:38:30
# 3: 115523231209 2017-01-01 12:21:51 13:17:38  19.033333 mins 00:19:02
# 4: 115523231209 2017-01-01 13:36:40 13:46:19 950.916667 mins 15:50:55
# 5: 115526742250 2017-01-02 05:37:14 05:43:02  11.633333 mins 00:11:38
# 6: 115526742250 2017-01-02 05:54:40 06:36:28 592.866667 mins 09:52:52
# 7: 115509240124 2017-01-02 16:29:20 17:26:50   3.566667 mins 00:03:34
# 8: 115509240124 2017-01-02 17:30:24 17:36:37 716.216667 mins 11:56:13
# 9: 115539253765 2017-01-03 05:32:50 05:40:51   1.600000 mins 00:01:36
# 10: 115539253765 2017-01-03 05:42:27 06:24:58 631.483333 mins 10:31:29
# 11: 115570245886 2017-01-03 16:56:27 18:23:17 199.133333 mins 03:19:08
# 12: 115567046025 2017-01-03 21:42:25 22:15:05 467.683333 mins 07:47:41
# 13: 115562452408 2017-01-04 06:02:46 06:32:15   8.283333 mins 00:08:17
# 14: 115562452408 2017-01-04 06:40:32 08:17:58 534.283333 mins 08:54:17
# 15: 115574565032 2017-01-04 17:12:15 18:14:35   1.833333 mins 00:01:50
# 16: 115574565032 2017-01-04 18:16:25 18:25:18         NA mins       NA

解决方案3
 1  2017-08-18 11:30:14

You need to make the times date-aware and shift one column to match the corresponding entries. 您需要使时间知道日期，并移动一列以匹配相应的条目。 For example: 例如： 

library(dplyr) # Needed for lead function

# Consider 'x' as your DF

x$entDate <- as.POSIXct(paste(x$BIZ_DT, x$ENTRY_TM)) # make entry date-aware
x$extDate <- as.POSIXct(paste(x$BIZ_DT, x$EXIT_TM)) # make exit date-aware

x$Stay_Duration <- as.numeric(lead(x$entDate, 1) - x$extDate, units = "secs") # Timedelta in seconds, could also be 'mins' etc.

#     JRNY_ID_NUM  BIZ_DT     ENTRY_TM EXIT_TM  entDate             extDate                     Stay_Duration
# 160 115492027250 2017-01-01 06:45:55 06:54:50 2017-01-01 06:45:55 2017-01-01 06:54:50          2386
# 73  115492027250 2017-01-01 07:34:36 07:43:21 2017-01-01 07:34:36 2017-01-01 07:43:21         16710
# 51  115523231209 2017-01-01 12:21:51 13:17:38 2017-01-01 12:21:51 2017-01-01 13:17:38          1142
# 145 115523231209 2017-01-01 13:36:40 13:46:19 2017-01-01 13:36:40 2017-01-01 13:46:19         57055
# 111 115526742250 2017-01-02 05:37:14 05:43:02 2017-01-02 05:37:14 2017-01-02 05:43:02           698
# 56  115526742250 2017-01-02 05:54:40 06:36:28 2017-01-02 05:54:40 2017-01-02 06:36:28         35572
# 119 115509240124 2017-01-02 16:29:20 17:26:50 2017-01-02 16:29:20 2017-01-02 17:26:50           214
# 157 115509240124 2017-01-02 17:30:24 17:36:37 2017-01-02 17:30:24 2017-01-02 17:36:37         42973
# 168 115539253765 2017-01-03 05:32:50 05:40:51 2017-01-03 05:32:50 2017-01-03 05:40:51            96
# 131 115539253765 2017-01-03 05:42:27 06:24:58 2017-01-03 05:42:27 2017-01-03 06:24:58         37889
# 81  115570245886 2017-01-03 16:56:27 18:23:17 2017-01-03 16:56:27 2017-01-03 18:23:17         11948
# 78  115567046025 2017-01-03 21:42:25 22:15:05 2017-01-03 21:42:25 2017-01-03 22:15:05         28061
# 135 115562452408 2017-01-04 06:02:46 06:32:15 2017-01-04 06:02:46 2017-01-04 06:32:15           497
# 35  115562452408 2017-01-04 06:40:32 08:17:58 2017-01-04 06:40:32 2017-01-04 08:17:58         32057
# 165 115574565032 2017-01-04 17:12:15 18:14:35 2017-01-04 17:12:15 2017-01-04 18:14:35           110
# 25  115574565032 2017-01-04 18:16:25 18:25:18 2017-01-04 18:16:25 2017-01-04 18:25:18            NA

But be careful, this solution does not apply to situations where entry and exit in the same row do occur in different dates (thus my comment). 但请注意，此解决方案不适用于同一行中的出入确实发生在不同日期的情况（因此，我的评论）。

解决方案4
 1  2017-08-18 11:36:41

Here is a naive base R possibility ( naive , because it is not aware of the date and simply uses the times): 这是幼稚的基于R的可能性（ naive ，因为它不知道日期，只是使用时间）： 

res <- df
res$Stay_Duration <-
  as.difftime(c(as.integer(difftime( 
    strptime(tail(res$ENTRY_TM, -1), "%H:%M:%S"),
    strptime(head(res$EXIT_TM, -1), "%H:%M:%S"), 
    units = "secs"
  )) %% (24*3600), NA), units = "secs")

#      JRNY_ID_NUM     BIZ_DT ENTRY_TM  EXIT_TM Stay_Duration
# 160 115492027250 2017-01-01 06:45:55 06:54:50     2386 secs
# 73  115492027250 2017-01-01 07:34:36 07:43:21    16710 secs
# 51  115523231209 2017-01-01 12:21:51 13:17:38     1142 secs
# 145 115523231209 2017-01-01 13:36:40 13:46:19    57055 secs
# 111 115526742250 2017-01-02 05:37:14 05:43:02      698 secs
# 56  115526742250 2017-01-02 05:54:40 06:36:28    35572 secs
# 119 115509240124 2017-01-02 16:29:20 17:26:50      214 secs
# 157 115509240124 2017-01-02 17:30:24 17:36:37    42973 secs
# 168 115539253765 2017-01-03 05:32:50 05:40:51       96 secs
# 131 115539253765 2017-01-03 05:42:27 06:24:58    37889 secs
# 81  115570245886 2017-01-03 16:56:27 18:23:17    11948 secs
# 78  115567046025 2017-01-03 21:42:25 22:15:05    28061 secs
# 135 115562452408 2017-01-04 06:02:46 06:32:15      497 secs
# 35  115562452408 2017-01-04 06:40:32 08:17:58    32057 secs
# 165 115574565032 2017-01-04 17:12:15 18:14:35      110 secs
# 25  115574565032 2017-01-04 18:16:25 18:25:18       NA secs

It simply compares the times, and applies a modulo operation. 它只是比较时间，并进行模运算。

As others have already said, you might run into different sorts of problems in case of gaps in dates. 正如其他人已经说过的那样，如果出现日期间隔，您可能会遇到不同类型的问题。

Consistency check with the desired output: 与所需输出的一致性检查： 

desired_output <-
  read.table(text =
"JRNY_ID_NUM     BIZ_DT      ENTRY_TM    EXIT_TM     Stay_Duration
115492027250    1/1/2017    6:45:55     6:54:50     0:39:46
115492027250    1/1/2017    7:34:36     7:43:21     4:38:30
115523231209    1/1/2017    12:21:51    13:17:38    0:19:02
115523231209    1/1/2017    13:36:40    13:46:19    15:50:55
115526742250    1/2/2017    5:37:14     5:43:02     0:11:38
115526742250    1/2/2017    5:54:40     6:36:28     9:52:52
115509240124    1/2/2017    16:29:20    17:26:50    0:03:34
115509240124    1/2/2017    17:30:24    17:36:37    11:56:13
115539253765    1/3/2017    5:32:50     5:40:51     0:01:36
115539253765    1/3/2017    5:42:27     6:24:58     10:31:29
115570245886    1/3/2017    16:56:27    18:23:17    3:19:08
115567046025    1/3/2017    21:42:25    22:15:05    7:47:41
115562452408    1/4/2017    6:02:46     6:32:15     0:08:17
115562452408    1/4/2017    6:40:32     8:17:58     8:54:17
115574565032    1/4/2017    17:12:15    18:14:35    0:01:50
115574565032    1/4/2017    18:16:25    18:25:18    place_holder",
  stringsAsFactors = FALSE, header = TRUE)

all.equal(
  as.difftime(desired_output$Stay_Duration, "%H:%M:%S", "secs"),
  res$Stay_Duration
)
# [1] TRUE

暂无
暂无

声明:本站的技术帖子网页，遵循CC BY-SA 4.0协议，如果您需要转载，请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

相关问题 行之间的时间差乘以因子R - difference Time between rows by factor R 计算R中后续行中两个值之间的差 - Calculating the difference between two values in subsequent rows in R 这两个命令在R中有什么区别？（因子变量的级别） - what's the difference between these two command in R?(levels of factor variable) 计算两列之间的时间差 - Calculating Time Difference between two columns NA 两行之间的时差 - Time difference between two rows with NA 取两个水平的因子变量之差，同时保留R中的其他因子变量 - Take difference between two levels of factor variable while retaining other factor variables in R 如何计算来自不同列的两个连续行之间的时间差？ - How can the time difference between two consecutive rows from different columns be calculated? 计算不同行的列之间的时间差（difftime） - Calculate time difference (difftime) between columns of different rows 如何计算R中可变行之间的时间差？ - How to calculate difference in time between variable rows in R? 从两个不同的列满足条件时检索后续行 - Retrieve subsequent rows when condition is met from two different columns
相关标签
 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM