嵌套循环输出到data.frame

时间:2015-01-13 03:34:32

标签: r loops nested-loops

我在R中有两个数据集(下面这些表只是较小的版本),我想将它们组合成一个新的数据框。

> meetingtime2     
#two columns of datetime that class=factor

               ST                  ET
1 2014-12-22 07:00:00 2014-12-22 07:30:00
2 2014-12-22 07:30:00 2014-12-22 08:00:00
3 2014-12-22 08:00:00 2014-12-22 08:30:00
4 2014-12-22 08:30:00 2014-12-22 09:00:00
5 2014-12-22 09:00:00 2014-12-22 09:30:00

> roomdata2 
#three columns; Room=factor, Capacity=integer, Video Conference=numeric

   Room Capacity Video.Conference
1 0M02A       16                1
2 0M03A        8                0
3 0M03B       12                1

所需的输出是15行×5列矩阵。简单来说,每个房间的每个时段都有输出。

#the following is a MANUALLY created output of what the first few rows should look like

    Room Capacity Video.Conference        ST                ET
 1 0M02A   16           1       2014-12-22 07:00:00 2014-12-22 07:30:00
 2 0M02A   16           1       2014-12-22 07:30:00 2014-12-22 08:00:00
 3 0M02A   16           1       2014-12-22 08:00:00 2014-12-22 08:30:00
 4 0M02A   16           1       2014-12-22 08:30:00 2014-12-22 09:00:00
 5 0M02A   16           1       2014-12-22 09:00:00 2014-12-22 09:30:00
 6 0M03A   16           1       2014-12-22 07:00:00 2014-12-22 07:30:00
 7 0M03A   16           1       2014-12-22 07:30:00 2014-12-22 08:00:00
#and so forth to 15 rows. 

我尝试过使用嵌套循环

#note, the code is written so I can apply to a bigger (1000's of rows) dataset

 >mylist<-list() 
 >for(i in 1:(nrow(roomdata2)))   
   +{   for(j in 1:(nrow(meetingtime2)))   
 +mylist[[j]]<-      data.frame(roomdata2[i,1],roomdata2[i,2],roomdata2[i,3],
 +meetingtime2[j,1],meetingtime2[j,2])  
  } 
   >df<-do.call("rbind",mylist)  
>df 

我得到的输出。我得到了最后一个房间的所有时间段,而不是前面的房间

roomdata2.i..1. roomdata2.i..2. roomdata2.i..3.  meetingtime2.j..1.  meetingtime2.j..2.
1    0M03B          12             1         2014-12-22 07:00:00    2014-12-22 07:30:00
2    0M03B          12             1         2014-12-22 07:30:00    2014-12-22 08:00:00
3    0M03B          12             1         2014-12-22 08:00:00    2014-12-22 08:30:00
4    0M03B          12             1         2014-12-22 08:30:00    2014-12-22 09:00:00
5    0M03B          12             1         2014-12-22 09:00:00    2014-12-22 09:30:00

我知道我的代码远非正确,并且正在给我循环的最后一次迭代。

我看到的另一种方法是每次迭代的连续打印功能

 >for(i in 1:(nrow(roomdata2))) 
 >for(j in 1:(nrow(meetingtime2))) 
 >print(paste(roomdata2[i,1],roomdata2[i,2],roomdata2[i,3],
 +meetingtime2[j,1],meetingtime2[j,2]))

输出

 [1] "0M02A 16 1 2014-12-22 07:00:00 2014-12-22 07:30:00"
 [1] "0M02A 16 1 2014-12-22 07:30:00 2014-12-22 08:00:00"
 [1] "0M02A 16 1 2014-12-22 08:00:00 2014-12-22 08:30:00"
 [1] "0M02A 16 1 2014-12-22 08:30:00 2014-12-22 09:00:00"
 [1] "0M02A 16 1 2014-12-22 09:00:00 2014-12-22 09:30:00"
 [1] "0M03A 8 0 2014-12-22 07:00:00 2014-12-22 07:30:00"
 [1] "0M03A 8 0 2014-12-22 07:30:00 2014-12-22 08:00:00"
 [1] "0M03A 8 0 2014-12-22 08:00:00 2014-12-22 08:30:00"
 [1] "0M03A 8 0 2014-12-22 08:30:00 2014-12-22 09:00:00"
 [1] "0M03A 8 0 2014-12-22 09:00:00 2014-12-22 09:30:00"
 [1] "0M03B 12 1 2014-12-22 07:00:00 2014-12-22 07:30:00"
 [1] "0M03B 12 1 2014-12-22 07:30:00 2014-12-22 08:00:00"
 [1] "0M03B 12 1 2014-12-22 08:00:00 2014-12-22 08:30:00"
 [1] "0M03B 12 1 2014-12-22 08:30:00 2014-12-22 09:00:00"
 [1] "0M03B 12 1 2014-12-22 09:00:00 2014-12-22 09:30:00"

#however the values are not separated, they are just in one set of string for each row.

所需的结果是直接在上面的表格,而是一个数据框,其中每个值都在一个单独的列中(每个日期和时间在一列中设置在一起)。

我已经查看了列表,lapply,foreach但是我无法绕过解决方案。 任何帮助将不胜感激,我是一个初学者,所以我很想学习。

干杯 * dputs

>dput(meetingtime2)

结构(列表(ST =结构(1:5,.Label = c(&#34; 22/12/2014 7:00&#34;, &#34; 22/12/2014 7:30&#34;,&#34; 22/12/2014 8:00&#34;,&#34; 22/12/2014 8:30&#34;,&# 34; 22/12/2014 9:00&#34; ),class =&#34; factor&#34;),ET = structure(1:5,.Label = c(&#34; 22/12/2014 7:30&#34;, &#34; 22/12/2014 8:00&#34;,&#34; 22/12/2014 8:30&#34;,&#34; 22/12/2014 9:00&#34;,&# 34; 22/12/2014 9:30&#34; ),class =&#34; factor&#34;)),. Name = c(&#34; ST&#34;,&#34; ET&#34;),row.names = c(NA, -5L),class =&#34; data.frame&#34;)

>dput(roomdata2)

结构(列表(房间=结构(1:3,.Label = c(&#34; 0M02A&#34;,&#34; 0M03A&#34;, &#34; 0M03B&#34;),class =&#34; factor&#34;),Capacity = c(16L,8L,12L),Video.Conference = c(1L, 0L,1L)),。Name = c(&#34; Room&#34;,&#34; Capacity&#34;,&#34; Video.Conference&#34;),row.names = c(NA, -3L),class =&#34; data.frame&#34;)

2 个答案:

答案 0 :(得分:3)

使用您的数据:

meetingtime2 <- read.csv(text = "ST,ET
2014-12-22 07:00:00,2014-12-22 07:30:00
2014-12-22 07:30:00,2014-12-22 08:00:00
2014-12-22 08:00:00,2014-12-22 08:30:00
2014-12-22 08:30:00,2014-12-22 09:00:00
2014-12-22 09:00:00,2014-12-22 09:30:00")

roomdata2 <- read.csv(text = "Room,Capacity,Video_Conference
0M02A,16,1
0M03A,8,0
0M03B,12,1")

然后merge方便地返回笛卡尔积,因为没有列匹配。

merge(meetingtime2, roomdata2)[, c(3:5, 1:2)]

##     Room Capacity Video_Conference                  ST                  ET
## 1  0M02A       16                1 2014-12-22 07:00:00 2014-12-22 07:30:00
## 2  0M02A       16                1 2014-12-22 07:30:00 2014-12-22 08:00:00
## 3  0M02A       16                1 2014-12-22 08:00:00 2014-12-22 08:30:00
## 4  0M02A       16                1 2014-12-22 08:30:00 2014-12-22 09:00:00
## 5  0M02A       16                1 2014-12-22 09:00:00 2014-12-22 09:30:00

答案 1 :(得分:0)

这很难看,但应该完成工作。鉴于以下数据:

ST <- c('2014-12-22 07:00:00', '2014-12-22 07:30:00', '2014-12-22 08:00:00', '2014-12-22 08:30:00', '2014-12-22 09:00:00')
ET <- c('2014-12-22 07:30:00', '2014-12-22 08:00:00', '2014-12-22 08:30:00', '2014-12-22 09:00:00', '2014-12-22 09:30:00')

RoomName <- c('0M02A', '0M03A', '0M03B')
Capacity <- c(16, 8, 12)
VideoCap <- c(1, 0, 1)

Times <- data.frame(ST, ET, stringsAsFactors = FALSE)
Rooms <- data.frame(RoomName, Capacity, VideoCap,stringsAsFactors = FALSE)

下面的功能应该做你想要的:

Smash <- function(DF1, DF2){
  nm <- dim(DF1)
  pq <- dim(DF2)    
  maxrow <- nm[[1]] * pq[[1]]
  maxcol <- nm[[2]] + pq[[2]]
  MAT <- matrix('A', nrow = maxrow, ncol = maxcol)
  currow <- 1
    for (i1 in seq_len(nm[[1]])) {
      for (i2 in seq_len(pq[[1]])) {
        curcol <- 1
        for (j in seq_len(nm[[2]])) {
          MAT[currow, curcol] <- DF1[i1, j]
          curcol <- curcol + 1
        }
        for (j in seq_len(pq[[2]])) {
          MAT[currow, curcol] <- DF2[i2, j]
          curcol <- curcol + 1
        }
        currow <- currow + 1
      }
    }
  DF <- data.frame(MAT)
  names(DF) <- c(names(DF1), names(DF2))
  return(DF)
}

Smash(Rooms,Times)返回:

> Smash(Rooms, Times)
   RoomName Capacity VideoCap                  ST                  ET
1     0M02A       16        1 2014-12-22 07:00:00 2014-12-22 07:30:00
2     0M02A       16        1 2014-12-22 07:30:00 2014-12-22 08:00:00
3     0M02A       16        1 2014-12-22 08:00:00 2014-12-22 08:30:00
4     0M02A       16        1 2014-12-22 08:30:00 2014-12-22 09:00:00
5     0M02A       16        1 2014-12-22 09:00:00 2014-12-22 09:30:00
6     0M03A        8        0 2014-12-22 07:00:00 2014-12-22 07:30:00
7     0M03A        8        0 2014-12-22 07:30:00 2014-12-22 08:00:00
8     0M03A        8        0 2014-12-22 08:00:00 2014-12-22 08:30:00
9     0M03A        8        0 2014-12-22 08:30:00 2014-12-22 09:00:00
10    0M03A        8        0 2014-12-22 09:00:00 2014-12-22 09:30:00
11    0M03B       12        1 2014-12-22 07:00:00 2014-12-22 07:30:00
12    0M03B       12        1 2014-12-22 07:30:00 2014-12-22 08:00:00
13    0M03B       12        1 2014-12-22 08:00:00 2014-12-22 08:30:00
14    0M03B       12        1 2014-12-22 08:30:00 2014-12-22 09:00:00
15    0M03B       12        1 2014-12-22 09:00:00 2014-12-22 09:30:00