R Shiny更改daterange显示周期后coefficient值异常修复求助

问题核心原因 #

return_coef函数计算系数时需要依赖全量原始数据的所有行、所有DR系列字段做统计计算。你当前代码中在data_subset响应式逻辑里，先按选择的日期范围过滤了数据集，再把过滤后的子集传入return_coef做计算，统计基准随着选择的日期范围变化，才会导致同一日期在不同选择范围下系数值不一致。

修复方法 #

仅修改server中data_subset部分的逻辑：日期范围过滤仅用来筛选最终要展示的行，计算系数时仍然传入完整的原始数据集即可。同时补充修正原代码中正则表达式未转义、分组残留警告的语法问题。

修复后完整可运行代码 #

library(shiny)
library(shinythemes)
library(dplyr)
library(writexl)
library(tidyverse)
library(lubridate)

function.test<-function(){
  
  df1 <- structure(
    list(date1= c("2021-06-28","2021-06-28","2021-06-28"),
         date2 = c("2021-07-01","2021-07-02","2021-07-04"),
         Category = c("ABC","ABC","ABC"),
         Week= c("Wednesday","Wednesday","Wednesday"),
         DR1 = c(4,1,0),
         DR01 = c(4,1,0), DR02= c(4,2,0),DR03= c(9,5,0),
         DR04 = c(5,4,0),DR05 = c(5,4,0),DR06 = c(5,4,0),DR07 = c(5,4,0),DR08 = c(5,4,0)),
    class = "data.frame", row.names = c(NA, -3L))
  
  return(df1)
  
}

return_coef <- function(df1, dmda, CategoryChosse) {
  
  x<-df1 %>% select(starts_with("DR0"))
  
  x<-cbind(df1, setNames(df1$DR1 - x, paste0(names(x), "_PV")))
  PV<-select(x, date2,Week, Category, DR1, ends_with("PV"))
  
  med<-PV %>%
    group_by(Category,Week) %>%
    summarize(across(ends_with("PV"), median), .groups = "drop")
  
  SPV<-df1%>%
    inner_join(med, by = c('Category', 'Week')) %>%
    mutate(across(matches("^DR0\\d+$"), ~.x + 
                    get(paste0(cur_column(), '_PV')),
                  .names = '{col}_{col}_PV')) %>%
    select(date1:Category, DR01_DR01_PV:last_col())
  
  SPV<-data.frame(SPV)
  
  mat1 <- df1 %>%
    filter(date2 == dmda, Category == CategoryChosse) %>%
    select(starts_with("DR0")) %>%
    pivot_longer(cols = everything()) %>%
    arrange(desc(row_number())) %>%
    mutate(cs = cumsum(value)) %>%
    filter(cs == 0) %>%
    pull(name)
  
  (dropnames <- paste0(mat1,"_",mat1, "_PV"))
  
  SPV <- SPV %>%
    filter(date2 == dmda, Category == CategoryChosse) %>%
    select(-any_of(dropnames))
  
  if(length(grep("DR0", names(SPV))) == 0) {
    SPV[head(mat1,10)] <- NA_real_
  }
  
  datas <-SPV %>%
    filter(date2 == ymd(dmda)) %>%
    group_by(Category) %>%
    summarize(across(starts_with("DR0"), sum), .groups = "drop") %>%
    pivot_longer(cols= -Category, names_pattern = "DR0(.+)", values_to = "val") %>%
    mutate(name = readr::parse_number(name))
  colnames(datas)[-1]<-c("Days","Numbers")
  
  
  datas <- datas %>% 
    group_by(Category) %>% 
    slice((as.Date(dmda) - min(as.Date(df1$date1) [
      df1$Category == first(Category)])):max(Days)+1) %>%
    ungroup
  
  m<-df1 %>%
    group_by(Category,Week) %>%
    summarize(across(starts_with("DR1"), mean), .groups = "drop")
  
  m<-subset(m, Week == df1$Week[match(ymd(dmda), ymd(df1$date2))] & Category == CategoryChosse)$DR1
  
  
  if (nrow(datas)<=2){
    return (as.numeric(m))
  }
  
  else if(any(table(datas$Numbers) >= 3) & length(unique(datas$Numbers)) == 1){
    yz <- unique(datas$Numbers)
    return(as.numeric(yz))
    
  }
  
  else{
    mod <- nls(Numbers ~ b1*Days^2+b2,start = list(b1 = 0,b2 = 0),data = datas, algorithm = "port")
    return(as.numeric(coef(mod)[2]))
  }
  
}

ui <- fluidPage(
  
  shiny::navbarPage(theme = shinytheme("flatly"), collapsible = TRUE,
                    br(),
                    tabPanel("",
                             sidebarLayout(
                               sidebarPanel(
                                 uiOutput('daterange'),
                                 br()
                                 
                               ),
                               mainPanel(
                                 dataTableOutput('table'),
                                 br(), br(),
                                 downloadButton("dl", "Download")
                               ),
                             ))
  ))

server <- function(input, output,session) {
  
  data <- reactive(function.test())
  
  data_subset <- reactive({
    req(input$daterange1)
    days <- seq(input$daterange1[1], input$daterange1[2], by = 'day')
    # 仅筛选要展示的行，不修改原始数据集
    df_show <- subset(data(), as.Date(date2) %in% days)
    df2 <- df_show %>% select(date2,Category)
    # 计算系数时传入完整原始数据集data()，而非过滤后的子集
    Test <- cbind(df2, coef = apply(df2, 1, function(x) {return_coef(data(),x[1],x[2])}))
    Test
  })
  
  output$daterange <- renderUI({
    dateRangeInput("daterange1", "Period you want to see:",
                   start = min(data()$date2),
                   end   = max(data()$date2))
  })
  
  output$table <- renderDataTable({
    data_subset()
  })
  
  output$dl <- downloadHandler(
    filename = function() { "data.xlsx"},
    content = function(file) {
      writexl::write_xlsx(data_subset(), path = file)
    }
  )
}

shinyApp(ui = ui, server = server) 

内容的提问来源于stack exchange，提问作者Antonio