Plotly带按钮图表图例与数值异常问题咨询

问题根源 #

当你使用add_trace并指定color = ~Program时，Plotly会为每个不同的Program单独生成一条轨迹。你的数据包含10个Program，因此第一次add_trace(Application)会生成10条轨迹，第二次add_trace(Enrolment)又生成10条轨迹，总计20条轨迹。但你在按钮的visible参数里仅设置了c(F, T)或c(T, F)，这只控制了前2条轨迹的可见性，剩余18条轨迹的可见性未被明确配置，导致点击按钮时出现数据混合显示、图例混乱的问题。

修正后的代码 #

library(plotly)
library(RColorBrewer)

# 创建示例数据
set.seed(32)
ae_data <- data.frame(
  Program = rep(c("Biology", "Math", "History", "English", "Computer Science",
                  "Chemistry", "Business", "Philosophy", "Psychology", "Physics"), 5),
  Year = rep(2018:2022, each = 10),
  Application = round(runif(50, min = 100, max = 900), digits = 0),
  Enrolment = round(runif(50, min = 150, max = 950), digits = 0)
)

# 生成对应数量的颜色
graph_colors <- colorRampPalette(brewer.pal(8, "Set2"))(length(unique(ae_data$Program)))

# 计算Program数量，生成完整的可见性控制向量
num_programs <- length(unique(ae_data$Program))
app_visible <- c(rep(TRUE, num_programs), rep(FALSE, num_programs))
enrol_visible <- c(rep(FALSE, num_programs), rep(TRUE, num_programs))

# 创建可切换图表
ae_data %>%
  plot_ly(
    type = "scatter",
    mode = "lines + markers",
    colors = graph_colors
  ) %>%
  add_trace(
    x = ~Year,
    y = ~Application,
    color = ~Program
  ) %>%
  add_trace(
    x = ~Year,
    y = ~Enrolment, 
    color = ~Program,
    visible = FALSE
  ) %>%
  layout(
    updatemenus = list(
      list(
        type = "buttons",
        buttons = list(
          list(
            label = "Application",
            method = "update",
            args = list(
              list(visible = app_visible),
              list(yaxis = list(title = "Application"))
            )
          ),
          list(
            label = "Enrolment",
            method = "update",
            args = list(
              list(visible = enrol_visible),
              list(yaxis = list(title = "Enrolment"))
            )
          )
        )
      )
    )
  )

核心修正说明 #

• 全轨迹可见性控制：通过num_programs获取Program数量，生成长度匹配所有轨迹的visible向量。切换到Application时，前10条（各Program的申请数据轨迹）设为可见，后10条设为隐藏；切换到Enrolment则反之。
• 颜色一致性保障：两次add_trace使用相同的颜色映射规则，确保同一Program在两个指标下的颜色一致，避免图例混淆。

内容的提问来源于stack exchange，提问作者user21505248