问

R正则表达式和聚合/子集

agree_6398026768 发布于 2023-02-09 22:44

在R中同时使用正则表达式时,是否可以聚合或使用子集？

我试图解决的问题是:我有一个名为'wpbCellFeatures'的数据框,其中包含多个列,包括唯一标识符'rowColFoVCell':

   rowColFoVCell wpbCount meanFeret meanPerim  meanCirc   meanAR meanRound meanSolidity
1   001001001001       38  1.182632  3.047368 0.7560526 1.948947 0.6036842    0.8289474
2   001001001002        8  1.886250  4.493750 0.7537500 2.365000 0.5350000    0.8325000

此列包含数字'001001001001','001001001002','001001001003',......,'001003004002',...等.形成此ID的数字对应于行号,列号,视野和单元格数字,例如'001003004002'是第一行,第三列,第四视野和第二单元格.

我想选择具有1到3之间的行的所有标识符,并聚合到新的数据框中.如何在R中执行此操作,我认为它将涉及使用聚合和正则表达式,但我对此不太熟悉？

谢谢

1 个回答

我不会使用正则表达式,而是通过使用read.fwf(或substr相关函数)将第一列拆分为相应的列.然后,将其绑定回原始数据集,并aggregate像往常一样使用等等.

toBind <- read.fwf(file = textConnection(as.character(mydf$rowColFoVCell)), 
                   widths = c(3, 3, 3, 3), colClasses = "character", 
                   col.names = c("Row", "Col", "FoV", "Cell"))
cbind(toBind, mydf)
#   Row Col FoV Cell rowColFoVCell wpbCount meanFeret meanPerim  meanCirc   meanAR meanRound
# 1 001 001 001  001  001001001001       38  1.182632  3.047368 0.7560526 1.948947 0.6036842
# 2 001 001 001  002  001001001002        8  1.886250  4.493750 0.7537500 2.365000 0.5350000
#   meanSolidity
# 1    0.8289474
# 2    0.8325000

在这里,我从"mydf"开始:

mydf <- structure(list(rowColFoVCell = c("001001001001", "001001001002"), 
                  wpbCount = c(38L, 8L), meanFeret = c(1.182632, 1.88625), 
                  meanPerim = c(3.047368, 4.49375), 
                  meanCirc = c(0.7560526, 0.75375), 
                  meanAR = c(1.948947, 2.365), 
                  meanRound = c(0.6036842, 0.535), 
                  meanSolidity = c(0.8289474, 0.8325)), 
                  .Names = c("rowColFoVCell", "wpbCount", "meanFeret", 
                             "meanPerim", "meanCirc", "meanAR", "meanRound",
                             "meanSolidity"), 
                  class = "data.frame", row.names = c(NA, -2L))

2023-02-09 23:06 回答

一米静心的阳光

撰写答案

今天，你开发时遇到什么问题呢？

立即提问

热门标签