R:5分钟间隔添加到数据帧

对意图的快速概述

• 传感器：触发传感器时的发生时间和房间（pings）
• 时隙：每个房间30分钟的结构化时间窗口，用于将数据排序到 查找每个房间每个5分钟区块中的ping数量。结构转换为所需的表

注意：

-每个30分钟的时间段（0-5分钟、5-10分钟、10-15分钟等）有6组5分钟窗口

-每个房间在ping计数方面与其他房间相同

-这是一个更大的装置的小版本（更多的房间，更多的30分钟街区）

-如果值为上午10:05:00，则属于括号“0到5分钟”，而不是“5-10”。 -10.05:01属于“5到10分钟”

想法： 如果传感器$Room=时隙$Room和时隙$ST<传感器$UTC=<时隙$ET ..不知何故，有一个计数指定它所属的5min块。在每次迭代和列计数时运行“sensor”的每一行。 我以前尝试过使用聚合函数，但很难填充“期望”输出的刚性结构

这远远超出了我的能力范围，因此任何帮助都将不胜感激

****添加了DPUT

>dput(timeslot)

structure(list(Room = structure(c(1L, 1L, 2L, 2L, 3L, 3L), .Label = c("Alpha", 
"Beta", "Gamma"), class = "factor"), Capacity = c(16L, 16L, 16L, 
16L, 10L, 10L), Video.Conference = c(1L, 1L, 1L, 1L, 0L, 0L), 
ST = structure(c(1L, 2L, 1L, 2L, 1L, 2L), .Label = c("2014-12-22 10:00:00", 
"2014-12-22 10:30:00"), class = "factor"), ET = structure(c(1L, 
2L, 1L, 2L, 1L, 2L), .Label = c("2014-12-22 10:30:00", "2014-12-22 11:00:00"
), class = "factor")), .Names = c("Room", "Capacity", "Video.Conference", 
"ST", "ET"), row.names = c(NA, -6L), class = "data.frame")

> dput(sensor)

structure(list(UTC = structure(c(4L, 5L, 6L, 7L, 9L, 11L, 12L, 
13L, 14L, 15L, 16L, 17L, 20L, 1L, 3L, 6L, 8L, 9L, 10L, 11L, 12L, 
13L, 14L, 15L, 18L, 19L, 21L, 2L, 22L), .Label = c("2014-12-22 10:08:34", 
"2014-12-22 10:09:34", "2014-12-22 10:14:00", "2014-12-22 10:14:34", 
"2014-12-22 10:15:00", "2014-12-22 10:15:25", "2014-12-22 10:15:27", 
"2014-12-22 10:15:29", "2014-12-22 10:16:00", "2014-12-22 10:17:00", 
"2014-12-22 10:18:00", "2014-12-22 10:18:03", "2014-12-22 10:18:10", 
"2014-12-22 10:18:39", "2014-12-22 10:19:00", "2014-12-22 10:19:04", 
"2014-12-22 10:19:06", "2014-12-22 10:19:09", "2014-12-22 10:19:11", 
"2014-12-22 10:20:00", "2014-12-22 10:20:20", "2014-12-22 10:39:11"
), class = "factor"), pin = c(1L, 0L, 0L, 1L, 1L, 0L, 0L, 1L, 
0L, 0L, 0L, 1L, 1L, 1L, 0L, 0L, 1L, 1L, 1L, 0L, 0L, 1L, 0L, 1L, 
0L, 1L, 0L, 1L, 1L), Room = structure(c(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, 3L, 3L), .Label = c("Alpha", "Beta", "Gamma"
), class = "factor")), .Names = c("UTC", "pin", "Room"), row.names = c(NA, 
-29L), class = "data.frame")

---

这里有一个方法。我相信有更好的方法来做这项工作。但是，这可能是您可以为更大的数据集进一步安排的内容

library(dplyr) # I use the dev version, dplyr 0.4
library(tidyr)
library(chron)

### Arrange sensor
sensor <- mutate(sensor, UTC = as.POSIXct(UTC, format = "%Y-%m-%d %H:%M:%S"),
                         minute = as.numeric(format(UTC, "%M")),
                         minute = ifelse(minute >= 30, minute-30, minute),
                         sub.group = as.character(cut(minute,
                                                      breaks = c(0, 5, 10, 15, 20, 25, 30),
                                                      labels = c("0-5", "5-10", "10-15", "15-20", "20-25", "25-30"))),
                         time = times(as.character(format(UTC, "%H:%M:%S"))),
                         time = as.character(cut(time,
                                                 breaks = times(c("10:00:00", "10:30:00", "11:00:00")),
                                                 labels = c("a", "b"))),
                         Room = as.character(Room))

### Create a master df
master <- data.frame(Room = rep(c("Alpha", "Beta", "Gamma"), each = 6),
                     time = rep(c("a", "b"), each = 18),
                     sub.group = rep(c("0-5", "5-10", "10-15",
                                       "15-20", "20-25", "25-30"), times = 3),
                     stringsAsFactors = FALSE)

### Get an outcome
df1 <- full_join(summarise(group_by(sensor, Room,time, sub.group), total = sum(pin)),
                 master, by = c("Room","time", "sub.group")) %>%
       spread(sub.group, total, fill = 0)


### Arrange timeslot for join
timeslot <- mutate_each(timeslot,funs(as.POSIXct(., format = "%Y-%m-%d %H:%M:%S")), ST:ET) %>%
            mutate(time = as.character(format(ST, "%M")),
                   time = ifelse(time == "00", "a", "b"),
                   Room = as.character(Room))

### join again
final <- full_join(timeslot, df1, by = c("Room", "time")) %>%
         select(-time)

### Order columns
final <- final[, c(1:6, 11, 7:10)]

### Get row sum
mutate(final, total = rowSums(final[, c(6:11)]))

#   Room Capacity Video.Conference                  ST                  ET 0-5 5-10 10-15 15-20 20-25 25-30 total
#1 Alpha       16                1 2014-12-22 10:00:00 2014-12-22 10:30:00   0    0     2     4     0     0     6
#2 Alpha       16                1 2014-12-22 10:30:00 2014-12-22 11:00:00   0    0     0     0     0     0     0
#3  Beta       16                1 2014-12-22 10:00:00 2014-12-22 10:30:00   0    1     1     5     0     0     7
#4  Beta       16                1 2014-12-22 10:30:00 2014-12-22 11:00:00   0    0     0     0     0     0     0
#5 Gamma       10                0 2014-12-22 10:00:00 2014-12-22 10:30:00   0    1     0     0     0     0     1
#6 Gamma       10                0 2014-12-22 10:30:00 2014-12-22 11:00:00   0    1     0     0     0     0     1

library（dplyr）#我使用开发版本dplyr 0.4
图书馆（tidyr）
图书馆（计时）
###排列传感器
传感器=30，分钟-30，分钟），
子组=作为字符（切割（分钟），
断裂=c（0,5,10,15,20,25,30），
标签=c（“0-5”、“5-10”、“10-15”、“15-20”、“20-25”、“25-30”），
时间=时间（以.character格式（UTC，%H:%M:%S）），
时间=作为字符（剪切（时间，
休息时间=时间（c（“10:00:00”、“10:30:00”、“11:00:00”），
标签=c（“a”、“b”），
房间=作为角色（房间））
###创建主df
主人
library(dplyr) # I use the dev version, dplyr 0.4
library(tidyr)
library(chron)

### Arrange sensor
sensor <- mutate(sensor, UTC = as.POSIXct(UTC, format = "%Y-%m-%d %H:%M:%S"),
                         minute = as.numeric(format(UTC, "%M")),
                         minute = ifelse(minute >= 30, minute-30, minute),
                         sub.group = as.character(cut(minute,
                                                      breaks = c(0, 5, 10, 15, 20, 25, 30),
                                                      labels = c("0-5", "5-10", "10-15", "15-20", "20-25", "25-30"))),
                         time = times(as.character(format(UTC, "%H:%M:%S"))),
                         time = as.character(cut(time,
                                                 breaks = times(c("10:00:00", "10:30:00", "11:00:00")),
                                                 labels = c("a", "b"))),
                         Room = as.character(Room))

### Create a master df
master <- data.frame(Room = rep(c("Alpha", "Beta", "Gamma"), each = 6),
                     time = rep(c("a", "b"), each = 18),
                     sub.group = rep(c("0-5", "5-10", "10-15",
                                       "15-20", "20-25", "25-30"), times = 3),
                     stringsAsFactors = FALSE)

### Get an outcome
df1 <- full_join(summarise(group_by(sensor, Room,time, sub.group), total = sum(pin)),
                 master, by = c("Room","time", "sub.group")) %>%
       spread(sub.group, total, fill = 0)


### Arrange timeslot for join
timeslot <- mutate_each(timeslot,funs(as.POSIXct(., format = "%Y-%m-%d %H:%M:%S")), ST:ET) %>%
            mutate(time = as.character(format(ST, "%M")),
                   time = ifelse(time == "00", "a", "b"),
                   Room = as.character(Room))

### join again
final <- full_join(timeslot, df1, by = c("Room", "time")) %>%
         select(-time)

### Order columns
final <- final[, c(1:6, 11, 7:10)]

### Get row sum
mutate(final, total = rowSums(final[, c(6:11)]))

#   Room Capacity Video.Conference                  ST                  ET 0-5 5-10 10-15 15-20 20-25 25-30 total
#1 Alpha       16                1 2014-12-22 10:00:00 2014-12-22 10:30:00   0    0     2     4     0     0     6
#2 Alpha       16                1 2014-12-22 10:30:00 2014-12-22 11:00:00   0    0     0     0     0     0     0
#3  Beta       16                1 2014-12-22 10:00:00 2014-12-22 10:30:00   0    1     1     5     0     0     7
#4  Beta       16                1 2014-12-22 10:30:00 2014-12-22 11:00:00   0    0     0     0     0     0     0
#5 Gamma       10                0 2014-12-22 10:00:00 2014-12-22 10:30:00   0    1     0     0     0     0     1
#6 Gamma       10                0 2014-12-22 10:30:00 2014-12-22 11:00:00   0    1     0     0     0     0     1