Python 根据条件为新列分配唯一值

这是一个数据.表方法，使用120秒的阈值

library(data.table)

# Set threshold (in seconds)
threshold = 120

# Get the car identifier
result=melt(
  setDT(df)[,trip:=.I][df, on=.(`Dropoff location`=`Pickup location`), nomatch=0] %>% 
    .[between(`i.Pickup time`-`Dropoff time`,0,threshold),.(trip,i.trip)] %>% 
    .[,car:=.I],id.vars = "car",value.name="trip"
)[,variable:=NULL][df, on="trip"]

# add any other single-instance cars
result[is.na(car),car:=seq(max(result$car,na.rm=T)+1, length.out = result[is.na(car),.N])]

输出:

     car  trip         Pickup time        Dropoff time Pickup location Dropoff location
   <int> <int>              <POSc>              <POSc>          <char>           <char>
1:     1     1 2016-06-09 21:06:36 2016-06-09 21:13:08               A                B
2:     2     2 2016-06-09 21:13:31 2016-06-09 21:23:59               A                C
3:     1     3 2016-06-09 21:13:45 2016-06-09 21:26:29               B                C
4:     3     4 2016-06-09 21:15:33 2016-06-09 21:44:31               A                B
5:     2     5 2016-06-09 21:24:49 2016-06-09 21:39:29               C                D

输入:

structure(list(`Pickup time` = structure(c(1465506396, 1465506811, 
1465506825, 1465506933, 1465507489), class = c("POSIXct", "POSIXt"
), tzone = "UTC"), `Dropoff time` = structure(c(1465506788, 1465507439, 
1465507589, 1465508671, 1465508369), class = c("POSIXct", "POSIXt"
), tzone = "UTC"), `Pickup location` = c("A", "A", "B", "A", 
"C"), `Dropoff location` = c("B", "C", "C", "B", "D")), row.names = c(NA, 
-5L), class = "data.frame")