Question

我正在研究Kaggle语义分段任务，

在我的代码的测试部分，

library(shiny)
library(ggplot2)
library(dplyr)
library(tidyverse)


## Only run examples in interactive R sessions
if (interactive()) {

  ui <- fluidPage(

    # App title ----
    titlePanel("Uploading Files"),

    # Sidebar layout with input and output definitions ----
    sidebarLayout(

      # Sidebar panel for inputs ----
      sidebarPanel(

        # Input: Select a file ----
        fileInput("file1", "Choose CSV File",
                  multiple = TRUE,
                  accept = c("text/csv",
                             "text/comma-separated-values,text/plain",
                             ".csv")),

        # Horizontal line ----
        tags$hr(),

        # Input: Checkbox if file has header ----
        checkboxInput("header", "Header", TRUE),

        # Input: Select separator ----
        radioButtons("sep", "Separator",
                     choices = c(Comma = ",",
                                 Semicolon = ";",
                                 Tab = "\t"),
                     selected = ","),

        # Input: Select quotes ----
        radioButtons("quote", "Quote",
                     choices = c(None = "",
                                 "Double Quote" = '"',
                                 "Single Quote" = "'"),
                     selected = '"'),

        # Horizontal line ----
        tags$hr(),

        # Input: Select number of rows to display ----
        radioButtons("disp", "Display",
                     choices = c(Head = "head",
                                 All = "all"),
                     selected = "head"),



        # Include a Slider for Strata
        sliderInput("strata",
                    "strata:",
                    min = 1,
                    max = 20,
                    value = c(1,20),
                    step=1)

      ),  

      ########################## 

      # Main panel for displaying outputs ----
      mainPanel(

        # Output: Data file ----
        tableOutput("contents")

      )  
    )
  )

  ###


  server <- function(input, output, session) {

    mytable <- reactive({

      req(input$file1)

      df <- read.csv(input$file1$datapath,
                     header = input$header,
                     sep = input$sep,
                     quote = input$quote, stringsAsFactors = FALSE)

      print(df)
      df<-data.frame(df)

      df<- df %>% 
        filter(df$strata>=input$strata[1] & df$strata<=input$strata[2])

      print(df)

      if(input$disp == "head") {
        return(head(df))
      }
      else {
        return(df)
      }

    })

    output$contents <- renderTable({
      # Now do use (), since we are calling a value from a reactive.
      mytable()
    })

  }
  # Run the app ----
  shinyApp(ui, server)

}

当我做preds部分时，数组只是填充，我希望它是一个最大位置的数组我不知道出了什么问题。 model = model.eval() predictions =[] for data in testdataloader: data = t.autograd.Variable(data, volatile=True).cuda() output = model.forward(data) _,preds = t.max(output, 1, keepdim = True)部分效果很好，我附上了output

可视化的屏幕截图

对出现问题的任何消息都会非常有帮助。

感谢

Answer 1

假设您的数据格式为MiniBatch x Dim，您现在正在查看哪个小批量具有最高价值。如果您使用单个样本（MB = 1）对其进行测试，那么您将始终获得0作为答案。因此，您可能想尝试：

_,preds = t.max(output, 0, keepdim = False)

输出语义分割任务

1 个答案: