R中每秒处理时间序列数据
Process time series data for each second in R
我的数据如下所示：
# Data Sample
Time Price V1 Time2 V2
2016-06-20 05:09:44 2086.50 1 05:09:44.284670 -1
2016-06-20 05:09:45 2086.75 5 05:09:45.212413 1
2016-06-20 05:09:45 2086.75 10 05:09:45.212413 1
2016-06-20 05:09:45 2086.75 1 05:09:45.212413 1
2016-06-20 05:09:46 2086.75 1 05:09:46.745124 1
2016-06-20 05:09:46 2086.75 1 05:09:46.745124 1
2016-06-20 05:09:46 2086.75 1 05:09:46.819954 1
2016-06-20 05:09:49 2086.75 1 05:09:49.279392 1
2016-06-20 05:09:49 2086.75 1 05:09:49.279392 1
2016-06-20 05:09:49 2086.75 1 05:09:49.352346 1
2016-06-20 05:09:49 2086.50 2 05:09:49.964023 -1
2016-06-20 05:09:49 2086.50 1 05:09:49.964023 -1
2016-06-20 05:09:55 2086.50 1 05:09:55.343324 -1
2016-06-20 05:09:57 2086.75 1 05:09:57.551886 1
2016-06-20 05:09:57 2086.75 1 05:09:57.650549 1
2016-06-20 05:09:57 2086.75 1 05:09:57.654352 1
2016-06-20 05:09:57 2086.75 1 05:09:57.654352 1
2016-06-20 05:09:57 2086.75 1 05:09:57.726578 1
我想清理数据，以便在每一秒内对所有 V1 求和。
所以我想要的输出看起来像：
# Desired Example
Time V1
2016-06-20 05:09:44 1
2016-06-20 05:09:45 16
2016-06-20 05:09:46 3
2016-06-20 05:09:47 0
2016-06-20 05:09:48 0
2016-06-20 05:09:49 6
2016-06-20 05:09:50 0
2016-06-20 05:09:51 0
2016-06-20 05:09:52 0
2016-06-20 05:09:53 0
2016-06-20 05:09:54 0
2016-06-20 05:09:55 1
2016-06-20 05:09:56 0
2016-06-20 05:09:57 5
我将第 "Time" 列转换为字符并将它们拆分并在列表中处理它们。但是，数据很大，计算时间太长。有没有办法通过动物园中的某些功能来做到这一点？
下面是使用 dput 的类似数据集：
结构(列表(V3 = c(2086.5, 2086.75, 2086.75, 2086.75, 2086.75,
2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75,
2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75,
2086.5, 2086.5, 2086.5, 2086.5, 2086.5, 2086.75, 2086.75, 2086.75,
2086.75, 2086.75, 2086.75, 2086.75, 2086.5, 2086.5, 2086.5, 2086.5,
2086.5, 2086.5, 2086.5, 2086.5, 2086.5, 2086.5, 2086.5, 2086.5,
2086.5, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75),
V4 = c(1L, 5L, 10L, 1L, 6L, 8L, 1L, 4L, 6L, 2L, 8L, 2L, 2L,
1L, 1L, 1L, 1L, 1L, 1L, 2L, 1L, 2L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 1L, 8L, 1L, 1L, 1L, 4L,
2L, 1L, 1L, 1L, 1L, 1L, 1L), V6 = c("05:09:44.284670", "05:09:45.212413",
"05:09:45.212413", "05:09:45.212413", "05:09:45.212413",
"05:09:45.299104", "05:09:45.299104", "05:09:45.301513",
"05:09:45.301513", "05:09:45.389110", "05:09:45.392840",
"05:09:45.475688", "05:09:45.543980", "05:09:46.745124",
"05:09:46.745124", "05:09:46.819954", "05:09:49.279392",
"05:09:49.279392", "05:09:49.352346", "05:09:49.964023",
"05:09:49.964023", "05:09:49.964023", "05:09:49.964023",
"05:09:55.343324", "05:09:57.551886", "05:09:57.650549",
"05:09:57.654352", "05:09:57.654352", "05:09:57.726578",
"05:09:57.728848", "05:09:58.286788", "05:10:00.390708",
"05:10:00.473617", "05:10:00.494903", "05:10:00.564042",
"05:10:08.24907", "05:10:09.633247", "05:10:09.633247", "05:10:09.633247",
"05:10:09.633247", "05:10:09.633247", "05:10:09.633247",
"05:10:09.633247", "05:10:09.633247", "05:10:09.830544",
"05:10:09.924001", "05:10:09.924001", "05:10:09.924001",
“05：10：09.924001”，“05：10：09.924001”），V7 = 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)), .Names = c("V3", "V4", "V6",
"V7"), row.names = c(NA, 50L), class = "data.frame")
data.table 非常快。尝试：
library(data.table)
library(lubridate)
mydata<-data.table(mydata)
mydata$Time<-ymd_hms(mydata$Time)
setkey(mydata, Time)
mydata.summed<-mydata[, .(V1 = sum(V1)), by = Time] # sums by each second
mydata2<-data.table(Time = seq(min(mydata$Time), max(mydata$Time), by = 1))
#create a new data.table to fill in the seconds you do not have values for
mydata<-mydata.summed[mydata2]
#merge them. see ?data.table for more information here
mydata[is.na(mydata)]<-0
#change the NAs that were created by the merge to 0
head(mydata, 10)
Time V1
1: 2016-06-20 05:09:44 1
2: 2016-06-20 05:09:45 16
3: 2016-06-20 05:09:46 3
4: 2016-06-20 05:09:47 0
5: 2016-06-20 05:09:48 0
6: 2016-06-20 05:09:49 6
7: 2016-06-20 05:09:50 0
8: 2016-06-20 05:09:51 0
9: 2016-06-20 05:09:52 0
10: 2016-06-20 05:09:53 0
我的数据如下所示：
# Data Sample
Time Price V1 Time2 V2
2016-06-20 05:09:44 2086.50 1 05:09:44.284670 -1
2016-06-20 05:09:45 2086.75 5 05:09:45.212413 1
2016-06-20 05:09:45 2086.75 10 05:09:45.212413 1
2016-06-20 05:09:45 2086.75 1 05:09:45.212413 1
2016-06-20 05:09:46 2086.75 1 05:09:46.745124 1
2016-06-20 05:09:46 2086.75 1 05:09:46.745124 1
2016-06-20 05:09:46 2086.75 1 05:09:46.819954 1
2016-06-20 05:09:49 2086.75 1 05:09:49.279392 1
2016-06-20 05:09:49 2086.75 1 05:09:49.279392 1
2016-06-20 05:09:49 2086.75 1 05:09:49.352346 1
2016-06-20 05:09:49 2086.50 2 05:09:49.964023 -1
2016-06-20 05:09:49 2086.50 1 05:09:49.964023 -1
2016-06-20 05:09:55 2086.50 1 05:09:55.343324 -1
2016-06-20 05:09:57 2086.75 1 05:09:57.551886 1
2016-06-20 05:09:57 2086.75 1 05:09:57.650549 1
2016-06-20 05:09:57 2086.75 1 05:09:57.654352 1
2016-06-20 05:09:57 2086.75 1 05:09:57.654352 1
2016-06-20 05:09:57 2086.75 1 05:09:57.726578 1
我想清理数据，以便在每一秒内对所有 V1 求和。 所以我想要的输出看起来像：
# Desired Example
Time V1
2016-06-20 05:09:44 1
2016-06-20 05:09:45 16
2016-06-20 05:09:46 3
2016-06-20 05:09:47 0
2016-06-20 05:09:48 0
2016-06-20 05:09:49 6
2016-06-20 05:09:50 0
2016-06-20 05:09:51 0
2016-06-20 05:09:52 0
2016-06-20 05:09:53 0
2016-06-20 05:09:54 0
2016-06-20 05:09:55 1
2016-06-20 05:09:56 0
2016-06-20 05:09:57 5
我将第 "Time" 列转换为字符并将它们拆分并在列表中处理它们。但是，数据很大，计算时间太长。有没有办法通过动物园中的某些功能来做到这一点？
下面是使用 dput 的类似数据集：
结构(列表(V3 = c(2086.5, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.5, 2086.5, 2086.5, 2086.5, 2086.5, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.5, 2086.5, 2086.5, 2086.5, 2086.5, 2086.5, 2086.5, 2086.5, 2086.5, 2086.5, 2086.5, 2086.5, 2086.5, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75, 2086.75), V4 = c(1L, 5L, 10L, 1L, 6L, 8L, 1L, 4L, 6L, 2L, 8L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 1L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 1L, 8L, 1L, 1L, 1L, 4L, 2L, 1L, 1L, 1L, 1L, 1L, 1L), V6 = c("05:09:44.284670", "05:09:45.212413", "05:09:45.212413", "05:09:45.212413", "05:09:45.212413", "05:09:45.299104", "05:09:45.299104", "05:09:45.301513", "05:09:45.301513", "05:09:45.389110", "05:09:45.392840", "05:09:45.475688", "05:09:45.543980", "05:09:46.745124", "05:09:46.745124", "05:09:46.819954", "05:09:49.279392", "05:09:49.279392", "05:09:49.352346", "05:09:49.964023", "05:09:49.964023", "05:09:49.964023", "05:09:49.964023", "05:09:55.343324", "05:09:57.551886", "05:09:57.650549", "05:09:57.654352", "05:09:57.654352", "05:09:57.726578", "05:09:57.728848", "05:09:58.286788", "05:10:00.390708", "05:10:00.473617", "05:10:00.494903", "05:10:00.564042", "05:10:08.24907", "05:10:09.633247", "05:10:09.633247", "05:10:09.633247", "05:10:09.633247", "05:10:09.633247", "05:10:09.633247", "05:10:09.633247", "05:10:09.633247", "05:10:09.830544", "05:10:09.924001", "05:10:09.924001", "05:10:09.924001", “05：10：09.924001”，“05：10：09.924001”），V7 = 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)), .Names = c("V3", "V4", "V6", "V7"), row.names = c(NA, 50L), class = "data.frame")
data.table 非常快。尝试：
library(data.table)
library(lubridate)
mydata<-data.table(mydata)
mydata$Time<-ymd_hms(mydata$Time)
setkey(mydata, Time)
mydata.summed<-mydata[, .(V1 = sum(V1)), by = Time] # sums by each second
mydata2<-data.table(Time = seq(min(mydata$Time), max(mydata$Time), by = 1))
#create a new data.table to fill in the seconds you do not have values for
mydata<-mydata.summed[mydata2]
#merge them. see ?data.table for more information here
mydata[is.na(mydata)]<-0
#change the NAs that were created by the merge to 0
head(mydata, 10)
Time V1
1: 2016-06-20 05:09:44 1
2: 2016-06-20 05:09:45 16
3: 2016-06-20 05:09:46 3
4: 2016-06-20 05:09:47 0
5: 2016-06-20 05:09:48 0
6: 2016-06-20 05:09:49 6
7: 2016-06-20 05:09:50 0
8: 2016-06-20 05:09:51 0
9: 2016-06-20 05:09:52 0
10: 2016-06-20 05:09:53 0