Question

我在解析数据帧中的数据时遇到问题。我正在使用igraph包来分析每天的模块化。我可以到达那个部分：

 Day_1 <- c("18", "34", NA, "25",NA, NA,NA,"39","18","29")
 Freq_1 <- c(0.6369427, 1.0615711, NA, 0.6369427,NA, NA,NA ,2.1231423, 0.7500849 ,2.7600849)
 Day_2 <- c("22", NA,"17", "14", "5" , "14", NA  , "6",  NA,   NA)
 Freq_2 <- c(2.5693731,NA, 23.0729702,  6.4234327,  1.8499486,  6.4234327,NA,  0.8221994, NA, NA)
 Day_3 <- c("16", NA ,  NA ,  "9",  "53", "17", "9"  ,"1",  "25", NA) 
 Freq_3 <- c(0.1573317, NA, NA, 5.6324733,0.9125236,13.2158590, 5.6324733,18.7853996 , 0.3461296,  NA)
 Day_4 <- c("74", "39", "14","35", "75", "59", "27", "47", "54", NA)
 Freq_4 <- c(0.1461988, 0.2192982, 0.2923977, 0.3654971, 0.4385965, 0.5116959, 0.8040936, 0.8771930, 0.9502924, NA)


test <- data.frame(Day_1,Freq_1,Day_2,Freq_2,Day_3,Freq_3,Day_4,Freq_4)
test

> test
   Day_1    Freq_1 Day_2     Freq_2 Day_3     Freq_3 Day_4    Freq_4
1     18 0.6369427    22  2.5693731    16  0.1573317    74 0.1461988
2     34 1.0615711  <NA>         NA  <NA>         NA    39 0.2192982
3   <NA>        NA    17 23.0729702  <NA>         NA    14 0.2923977
4     25 0.6369427    14  6.4234327     9  5.6324733    35 0.3654971
5   <NA>        NA     5  1.8499486    53  0.9125236    75 0.4385965
6   <NA>        NA    14  6.4234327    17 13.2158590    59 0.5116959
7   <NA>        NA  <NA>         NA     9  5.6324733    27 0.8040936
8     39 2.1231423     6  0.8221994     1 18.7853996    47 0.8771930
9     18 0.7500849  <NA>         NA    25  0.3461296    54 0.9502924
10    29 2.7600849  <NA>         NA  <NA>         NA  <NA>        NA

我需要在这些数据之间生成一个链接，以使用Sankey图生成视图。例如：

格式：day_X，day_X + 1，Freq_X + 1

 source <- c("Cluster 18A","Cluster 22B","Cluster 16C","Cluster 34A","Cluster 17B", "Cluster 25A","Cluster 14B","Cluster 9C","Cluster 5B", "Cluster 53C","Cluster 14B","Cluster 17C","Cluster 9C","Cluster 39A" ,"Cluster 6B","Cluster 1C", "Cluster 18A", "Cluster 25C")
    target <- c("Cluster 22B", "Cluster 16C","Cluster 74D","Cluster 39D","Cluster 14D","Cluster 14B","Cluster 9C","Cluster 35D", "Cluster 53C","Cluster 75D","Cluster 17C","Cluster 59D","Cluster 27D","Cluster 6B", "Cluster 1C","Cluster 47D", "Cluster 25C","Cluster 54D")
   value <- c(2.5693731,0.1573317,0.1461988,0.2192982,0.2923977,6.4234327,5.6324733, 0.3654971,0.9125236,0.4385965,13.2158590,0.5116959,0.8040936,0.8221994,18.7853996, 0.8771930,0.3461296 , 0.9502924)

    > test2 <- data.frame(source,target,value)
    > test2
            source      target      value
    1  Cluster 18A Cluster 22B  2.5693731
    2  Cluster 22B Cluster 16C  0.1573317
    3  Cluster 16C Cluster 74D  0.1461988
    4  Cluster 34A Cluster 39D  0.2192982
    5  Cluster 17B Cluster 14D  0.2923977
    6  Cluster 25A Cluster 14B  6.4234327
    7  Cluster 14B  Cluster 9C  5.6324733
    8   Cluster 9C Cluster 35D  0.3654971
    9   Cluster 5B Cluster 53C  0.9125236
    10 Cluster 53C Cluster 75D  0.4385965
    11 Cluster 14B Cluster 17C 13.2158590
    12 Cluster 17C Cluster 59D  0.5116959
    13  Cluster 9C Cluster 27D  0.8040936
    14 Cluster 39A  Cluster 6B  0.8221994
    15  Cluster 6B  Cluster 1C 18.7853996
    16  Cluster 1C Cluster 47D  0.8771930
    17 Cluster 18A Cluster 25C  0.3461296
    18 Cluster 25C Cluster 54D  0.9502924

我创建了这个函数，但它给我一些我无法解决的错误。

pased_data <- do.call(rbind,lapply(1:nrow(test), function(row){
  node <- test[row,]
  nomes <- names(node)
  node <- node[!is.na(node)]

  col_pares <- seq(4,length(node),2)
  col_impares <- seq(1,length(node),2)

  values <- as.numeric(node[col_pares])
  pairs <- paste0(as.character(node[col_impares]),
                  as.character(nomes[col_impares]))

  colsource <- pairs[1:(length(pairs)-1)]
  coltarget <- pairs[2:(length(pairs))]

  data <- data.frame(source = colsource,
                     target = coltarget,
                     value = values)
}) 
)

错误：

该功能仅适用于4天的分析;
当第二天的NA被删除时，day_3变为day_2;
如果我只有一天的数据，我不知道如何删除该行。（测试中的第10行）

我需要生成一个等于test2的数据帧，但是需要几天（取决于分析）并且没有错误。：）

如何按日解析数据并创建边列表

0 个答案: