Question

我的数据框看起来像：

table(df$partnumber, df$personworking)

等数千行。同一部分可以分配多个人。我总结 R 中的数据非常糟糕，但我能够通过键入来生成（在控制台中）一个看起来像频率矩阵的表：

Part Number   NumWorkers    Names
A             3             "James, Andrea"
B             1             "Brian"
C             1             "Tiffany"

它将独特的项目作为行，每个人的工作名称作为列进行吐出。值为0或1，具体取决于它们是否正在运行该部分。

我正在寻找的是一种以易于理解的格式汇总这些信息的方法，根据项目说明：

thedataframe <- data.frame(thetable[,])

我也在努力让我的桌子进入数据框架。我试过了：

    Dim fd As OpenFileDialog = New OpenFileDialog()

    fd.Title = "Open File Dialog"
    fd.InitialDirectory = "C:\"
    fd.Filter = "All files (*.xlsx)|*.xlsx|All files (*.xlsx)|*.xlsx"
    fd.FilterIndex = 2
    fd.RestoreDirectory = True
    Dim myStream As Stream = Nothing

    If fd.ShowDialog() = DialogResult.OK Then

        Try
            myStream = fd.OpenFile()
            If (myStream IsNot Nothing) Then
                workbook = APP.Workbooks.Open(fd.FileName)
                worksheet = workbook.Worksheets("sheet1")

                TextBox1.Text = worksheet.Cells(1, 7).Value
                TextBox2.Text = worksheet.Cells(2, 7).Value
                TextBox3.Text = worksheet.Cells(3, 7).Value

                l1.Text = worksheet.Cells(2, 1).Value * 10
                w1.Text = worksheet.Cells(2, 2).Value * 10
                q1.Text = worksheet.Cells(2, 3).Value
                p1.Text = worksheet.Cells(2, 4).Value..........

                Dim Values(119, 3) As String
                Values(0, 0) = l1.Text
                Values(0, 1) = w1.Text
                Values(0, 2) = q1.Text
                Values(0, 3) = p1.Text.........

                Dim add As Integer = 0
                Dim pressing As Integer = 0

                If adding.Text = "50" Or adding.Text = "" Then
                    add = 50
                Else
                    add = Convert.ToInt16(adding.Text)
                End If

                If press.Text = "20" Or press.Text = "" Then
                    pressing = 20
                Else
                    pressing = Convert.ToInt16(press.Text)
                End If

                Dim l As Integer
                Dim w As Integer
                Dim machinearea As Integer

                Dim connetionString As String
                Dim cnn As SqlConnection
                connetionString = "Data Source=.;Initial Catalog=lumber;User ID=sa;Password=sasql"
                cnn = New SqlConnection(connetionString)
                Dim cmd As SqlCommand
                Dim myreader As SqlDataReader
                Dim query As String

                query = "SELECT length,width from marea"
                cmd = New SqlCommand(query, cnn)
                cnn.Open()
                myreader = cmd.ExecuteReader()
                If myreader.Read() Then
                    l = myreader.Item("length")
                    w = myreader.Item("width")
                End If
                cnn.Close()

                machinearea = l * w

                Dim allTextBoxes2 = From txt In Me.Panel1.Controls.OfType(Of TextBox)()
           Order By txt.TabIndex
                Dim txtList2 = allTextBoxes2.ToList()

                For i As Int32 = 0 To txtList2.Count - 1
                    Dim thisTxt = txtList2(i)
                    Dim nextIndex = If(i + 1 >= txtList2.Count, 0, i + 1)
                    Dim prevIndex = If(i - 1 < 0, txtList2.Count - 1, i - 1)
                    Dim nextTxt = txtList2(nextIndex)
                    Dim prevTxt = txtList2(prevIndex)
                    Dim testInt As Integer = 0

                    If thisTxt.Text = "0" Then
                        thisTxt.Clear()
                    End If

                Next

                Dim allTextBoxes = From txt In Me.Panel1.Controls.OfType(Of TextBox)()
                              Order By txt.TabIndex
                Dim txtList = allTextBoxes.ToList()


                For i As Int32 = 0 To txtList.Count - 1
                    Dim thisTxt = txtList(i)
                    For j = 2 To -1
                        For k = 1 To 4
                            thisTxt.Text = worksheet.Cells(i + j, k).Value
                        Next
                    Next
                Next
            End If

        Catch Ex As Exception

            MessageBox.Show("Cannot read file from disk. Original error: " & Ex.Message)

        Finally
            ' Check this again, since we need to make sure we didn't throw an exception on open.
            If (myStream IsNot Nothing) Then
                myStream.Close()
            End If
        End Try
    End If

但我没有走得太远。我想总结每个独特部分的工作量，并连接并打印每个列名称，其中有一个作为给定部分的值。

在Base R 中汇总此数据的最佳方法是什么？

Answer 1

我们可以使用data.table。转换＆＃39; data.frame＆＃39;到＆＃39; data.table＆＃39; （setDT(df)），按＆＃39; partnumber＆＃39;分组，获取行数（.N）和paste＆＃39; personworking＆＃39;在每个＆＃39;部分编号＆＃39;。

library(data.table)
setDT(df)[,.(NumWorkers = .N, Names = toString(personworking)) , by = partnumber]

或我们可以使用dplyr

library(dplyr)
df %>%
   group_by(partnumber) %>%
   summarise(NumWorkers = n(), Names = toString(personworking))

或使用base R

do.call(rbind, by(df, df$partnumber, FUN = function(x) 
   data.frame(NumWorkers = length(x$personworking), Names = toString(x$personworking))))

Answer 2

以下是您可以在基座R中使用aggregate：

的方法

dfAgg <- do.call(data.frame,
           aggregate(df$Person, list(df$Parts),
                     FUN=function(x) c(length(x), paste(x, collapse=", "))))
# add nicer names
names(dfAgg) <- c("Parts", "Count", "Person")

Aggregate允许您在组上运行功能。在这个例子中，我们正在运行一个函数，它返回个体计数（通过length）及其名称（通过paste）。

以下是我用来测试它的示例数据。

数据

set.seed(1234) df <- data.frame("Parts"=sample(LETTERS[1:3], 10, replace=T), "Person"=sample(c("James", "Brian", "Sam", "Tiff", "Sandy"), 10, replace=T), stringsAsFactors=F)

从看起来像矩阵的表中获取列名和频率

2 个答案: