主页 / coding / 问题 / 79119329
Accepted
Amanda Goldberg
Asked: 2024-10-24 02:51:13 +0800 CST

如何删除在 60 分钟内收集的组内的重复项？

  • 772

我有相机陷阱数据，我想删除可能重复的动物检测。我将间隔设置为 60 分钟（1 小时），这样如果在不到 60 分钟的时间内再次检测到同一物种的任何个体，则该个体是同一个体。我的数据是在多个区块收集的，每个区块内有多个站点。

#Data example:
Block<-c("a","a","a","a","a","b","b","b","b","b") #2 block
Site<-c("p1","p1","p2","p2","p2","p1","p1","p1","p2","p2") #two sites
Species<-c("c","c","c","c","e","d","d","c","c","c")
datetime<-c("2021-03-29 05:45:00","2021-03-29 06:40:00","2021-03-30 05:45:00","2021-03-30 07:45:00","2021-03-29 09:45:00","2021-03-29 05:45:00","2021-03-29 05:55:00","2021-03-29 08:45:00","2021-03-29 10:45:00","2021-03-30 10:59:00")
df<-data.frame(Block, Site, Species, datetime)

#what I want for the output: 

   Block Site Species            datetime
1      a   p1       c 2021-03-29 05:45:00
2      a   p2       c 2021-03-30 05:45:00
3      a   p2       c 2021-03-30 07:45:00
4      a   p2       e 2021-03-29 09:45:00
5      b   p1       d 2021-03-29 05:45:00
6      b   p1       c 2021-03-29 08:45:00
7      b   p2       c 2021-03-29 10:45:00
8      b   p2       c 2021-03-30 10:59:00

至少对我来说，棘手的部分是我需要删除在 1 小时内出现的 Block/Site/Species 的重复项（我不能只说保留一天中一小时内的第一个）。

感谢您的帮助。

1 个回答

  1. Best Answer

    使用diff然后filter设置intv

    library(dplyr)

intv <- 60

df %>% 
  mutate(datetime = as.POSIXct(datetime)) %>% 
  arrange(Block, Site, Species, datetime) %>% 
  filter(c(intv + 1, diff(datetime, units="mins")) > intv, 
         .by = c(Block, Site, Species))
  Block Site Species            datetime
1     a   p1       c 2021-03-29 05:45:00
2     a   p2       c 2021-03-30 05:45:00
3     a   p2       c 2021-03-30 07:45:00
4     a   p2       e 2021-03-29 09:45:00
5     b   p1       c 2021-03-29 08:45:00
6     b   p1       d 2021-03-29 05:45:00
7     b   p2       c 2021-03-29 10:45:00
8     b   p2       c 2021-03-30 10:59:00
    • 1

相关问题