Question

我有一个棘手的打印输出格式，我试图获得。这是我的数据帧，它由for循环和rbind构建。

bets<- data.frame(status=character(), f_name=character(), d_name=character(), type_bet=character(), sec=character(),
                  spread=character(), total=character(), deriv=character(), book=character(), edge=character(), 
                  my_f_price=character(), book_f_price=character(), my_d_price=character(), book_d_price=character())

打印样本：

status  f_name  d_name  type_bet    sec spread  total   deriv   book    edge    my_f_price  book_f_price    my_d_price  book_d_price

9:00 PM ET  San Diego State Colorado State  total   h1  3.5 138.5   65  pin 12  120 -108    -120    -108
9:00 PM ET  San Diego State Colorado State  total   h1  3.5 138.5   65  5d  10  120 -110    -120    -110
6:00 PM ET  Cincinnati  SMU total   h1  8   125.5   59  pin 9   122 -103    -122    -113
8:00 PM ET  Temple  Rutgers total   h1  1.5 150 70.5    pin 8   116 -108    -116    -108
8:00 PM ET  Temple  Rutgers total   h1  1.5 150 70.5    5d  6   116 -110    -116    -110
8:05 PM ET  Drake   Evansville  ml  h1  7   136 0   5d  4   -214    -210    214 175
8:00 PM ET  Northern Iowa   Bradley total   h1  12  133 62  5d  3   113 -110    -113    -110
6:00 PM ET  Cincinnati  SMU ml  h1  8   125.5   0   5d  2   -242    -240    242 200
6:00 PM ET  Cincinnati  SMU total   h1  8   125.5   58.5    5d  2   112 -110    -112    -110

它有点难以看到，但边缘栏是如何订购的，12,10,9,8,6,4,3,2,2。我想做的是虽然将一些条目组合在一起当f_name，d_name，type_bet和sec都相同，并且唯一不同的列是书时，应将其视为一个组。理想情况下，我希望打印输出看起来像这样：

status  f_name  d_name  type_bet    sec spread  total   deriv   book    edge    my_f_price  book_f_price    my_d_price  book_d_price

9:00 PM ET  San Diego State Colorado State  total   h1  3.5 138.5   65  pin 12  120 -108    -120    -108
9:00 PM ET  San Diego State Colorado State  total   h1  3.5 138.5   65  5d  10  120 -110    -120    -110

6:00 PM ET  Cincinnati  SMU total   h1  8   125.5   59  pin 9   122 -103    -122    -113
6:00 PM ET  Cincinnati  SMU total   h1  8   125.5   58.5    5d  2   112 -110    -112    -110

8:00 PM ET  Temple  Rutgers total   h1  1.5 150 70.5    pin 8   116 -108    -116    -108
8:00 PM ET  Temple  Rutgers total   h1  1.5 150 70.5    5d  6   116 -110    -116    -110

8:05 PM ET  Drake   Evansville  ml  h1  7   136 0   5d  4   -214    -210    214 175

8:00 PM ET  Northern Iowa   Bradley total   h1  12  133 62  5d  3   113 -110    -113    -110

6:00 PM ET  Cincinnati  SMU ml  h1  8   125.5   0   5d  2   -242    -240    242 200

现在我能想到的唯一方法就是逐行打印到txt文件，循环遍历数据框（按边列排序），对于每个条目，我可以在数据框的其余部分搜索另一个条目同样的f_name，d_name，type_bet，sec和打印也是如此，并从数据帧中删除它。但我认为还有更好的方法吗？

Answer 1

我正在使用自己的数据框，因为它比处理上面的文本字符串要少。

假设您想要组成一个组的变量被称为formGroupVarX（在您的情况下为“f_name”，“d_name”，“type_bet”，“sec”）以及该FreeVarX之外的变量（所有其他变量）然后你可以显示如下：

formGroupVars = c("formGroupVar1","formGroupVar2","formGroupVar3")
freeVars = c("FreeVar1")
frameToShow <- data.frame(cbind(sample(LETTERS[1:3],20,replace=TRUE),sample(LETTERS[4:6],20,replace=TRUE),
                                sample(LETTERS[7:9],20,replace=TRUE),sample(letters,20,replace=TRUE)        ))
colnames(frameToShow) = c(formGroupVars,freeVars)
frameToShow[order(apply(frameToShow,1,function(X) {  paste(X[formGroupVars],collapse="")   } )),]

基本上，您创建一个临时因子级别，该级别由要组成组的所有变量的函数组成，并在该临时因子上排序显示。在你和我的例子中，一个简单的值连接可以解决这个问题，但理论上这个函数可以是一个数学函数或任何其他函数。

Answer 2

您的示例数据（您可以使用dput（yourData）来生成此示例 - 使其更容易提供帮助）

df <- structure(list(status = c("9:00 PM ET", "9:00 PM ET", "6:00 PM ET", 
 "8:00 PM ET", "8:00 PM ET", "8:05 PM ET", "8:00 PM ET", "6:00 PM ET", 
"6:00 PM ET"), f_name = c("San Diego State", "San Diego State", 
"Cincinnati", "Temple", "Temple", "Drake", "Northern Iowa", "Cincinnati", 
"Cincinnati"), 
 d_name = c("Colorado State", "Colorado State", "SMU", "Rutgers", "Rutgers", 
  "Evansville", "Bradley", "SMU", "SMU"), type_bet = c("total", "total", "total",      
"total", "total", 
 "ml", "total", "ml", "total"), sec = c("h1", "h1", "h1", "h1", 
 "h1", "h1", "h1", "h1", "h1"), spread = c(3.5, 3.5, 8, 1.5, 1.5, 
7, 12, 8, 8), total = c(138.5, 138.5, 125.5, 150, 150, 136, 133, 
125.5, 125.5), deriv = c(65, 65, 59, 70.5, 70.5, 0, 62, 0, 58.5), 
 book = c("pin", "5d", "pin", "pin", "5d", "5d", "5d", "5d", 
 "5d"), edge = c(12L, 10L, 9L, 8L, 6L, 4L, 3L, 2L, 2L), my_f_price = c(120L, 
 120L, 122L, 116L, 116L, -214L, 113L, -242L, 112L), book_f_price = c(-108L, 
-110L, -103L, -108L, -110L, -210L, -110L, -240L, -110L), my_d_price = c(-120L, 
-120L, -122L, -116L, -116L, 214L, -113L, 242L, -112L), book_d_price = c(-108L, 
 -110L, -113L, -108L, -110L, 175L, -110L, 200L, -110L)), .Names = c("status", 
"f_name", "d_name", "type_bet", "sec", "spread", "total", "deriv", 
"book", "edge", "my_f_price", "book_f_price", "my_d_price", "book_d_price" ),
              class = "data.frame", row.names = c(NA, -9L))

#You can sort your data on the required columns - but doesn't produce exactly the output you want
df2 <- df[order(df$f_name, df$d_name, df$type_bet, df$sec) , ]

不确定您想要的输出结构（即各组之间的空白是什么？），但你可以使用列表接近这一点。

#Split data by required groups (and remove empty dataframes produced by interaction)
df.grp <- split(df , list(df$f_name, df$d_name, df$type_bet, df$sec))
df.grp <- df.grp[sapply(df.grp, function(z) nrow(z)>0)]

#Get in the order of decreasing edge
max.edge <- unlist(lapply(df.grp , function(x) max(x[,'edge'])))
list.names <- names(sort(max.edge, decreasing=T))
(out <- df.grp[match(names(df.grp),list.names)])

R - 棘手的CSV打印输出

2 个答案: