如何从图中删除痕迹答案

【问题标题】：How to remove trace from the plot如何从图中删除痕迹
【发布时间】：2021-01-04 16:10:40
【问题描述】：

我想从图中和下面的图例中删除 NA 跟踪。

  plot_ly(data = bData,  x = ~`Maturity Date`, y =  ~YVal,  type = 'scatter', mode='markers',  
              symbol = ~Sym,  symbols = c('circle-open','x-open','diamond-open') ,
              text = ~paste(" Security: ", bData$Security, "<br>",
                            "Currency: ", bData$Crncy, "<br>",
                            "YTM: ", bData$YTM,"<br>",
                            "DM: ", bData$DM) ,
              hoverinfo = 'text',legendgroup = 'group1'
      ) %>%
        add_trace(x = ~`Maturity Date`, y =  ~YVal , symbol=~Crncy,legendgroup = 'group2')

【问题讨论】：

标签： r shiny plotly scatter-plot

【解决方案1】：

试试这个。您可以在新数据框中隔离没有任何值的数据，然后将其绘制出来。在此示例中，我将省略 Sym 变量中的 None 值，但如果您有 NA，请使用 !is.na()，因为我还为您添加了一行新代码。代码如下：

library(plotly)
#New data example
bData2 <- bData[bData$Sym!='None',]
#For your real data
bData2 <- bData[!is.na(bData$Sym),]
#Code
plot_ly(data = bData2,  x = ~`Maturity Date`, y =  ~YVal,  type = 'scatter', mode='markers',  
        symbol = ~Sym,  symbols = c('circle-open','x-open','diamond-open') ,
        text = ~paste(" Security: ", bData2$Security, "<br>",
                      "Currency: ", bData2$Crncy, "<br>",
                      "YTM: ", bData2$YTM,"<br>",
                      "DM: ", bData2$DM) ,
        hoverinfo = 'text',legendgroup = 'group1'
) %>%
  add_trace(x = ~`Maturity Date`, y =  ~YVal , symbol=~Crncy,legendgroup = 'group2')

输出省略None:

更新：试试这个技巧，将属于None 类的值替换为NA，首先将因子转换为字符，然后再返回因子。真正的NA 将被plotly 函数删除。代码如下：

#Replace
bData$Sym <- as.character(bData$Sym)
bData$Sym[bData$Sym=='None']<-NA
bData$Sym <- as.factor(bData$Sym)
#Code
plot_ly(data = bData,  x = ~`Maturity Date`, y =  ~YVal,  type = 'scatter', mode='markers',  
        symbol = ~Sym,  symbols = c('circle-open','x-open','diamond-open') ,
        text = ~paste(" Security: ", bData$Security, "<br>",
                      "Currency: ", bData$Crncy, "<br>",
                      "YTM: ", bData$YTM,"<br>",
                      "DM: ", bData$DM) ,
        hoverinfo = 'text',legendgroup = 'group1'
) %>%
  add_trace(x = ~`Maturity Date`, y =  ~YVal , symbol=~Crncy,legendgroup = 'group2')

输出：

或者也试试这个，复制YVal并为第一个变量设置NA：

#Replace 2
bData$YVal2 <- bData$YVal
bData$Sym <- as.character(bData$Sym)
bData$Sym[bData$Sym=='None']<-NA
bData$YVal[is.na(bData$Sym)]<-NA
bData$Sym <- as.factor(bData$Sym)
#Code
plot_ly(data = bData,  x = ~`Maturity Date`, y =  ~YVal,  type = 'scatter', mode='markers',  
        symbol = ~Sym,  symbols = c('circle-open','x-open','diamond-open') ,
        text = ~paste(" Security: ", bData$Security, "<br>",
                      "Currency: ", bData$Crncy, "<br>",
                      "YTM: ", bData$YTM,"<br>",
                      "DM: ", bData$DM) ,
        hoverinfo = 'text',legendgroup = 'group1'
) %>%
  add_trace(x = ~`Maturity Date`, y =  ~YVal2 , symbol=~Crncy,legendgroup = 'group2')

输出：

【讨论】：

实际上，对于第二次跟踪，数据源应该是 bData 而不是 bData2，因为 bData2 是 bData 的子集，它会错过一些在 bData$Sym == NA 中具有 NA 的货币
@AriKari 这是一个有效的观察结果，您可以尝试使用它或尝试na.omit() 或complete.cases()。您没有提供新数据，因此我无法正确重现该问题。如果你愿意，你可以分享你在新情节中使用的新数据！
@AriKari 如果您分享新数据，请告诉我！
新数据与我在之前的问题中分享的数据相同，除了它是 NA 而不是 None
@AriKari 我添加了一个更新，您能否检查一下是否适合您？