如何修改Plotly Treemap（树形图）的颜色？

要调整树形图的颜色，你只需要在plot_ly()函数里添加或修改相关参数即可，下面是几种实用的修改方式：

1. 自定义固定颜色数组 #

直接给每个类别指定固定颜色，在marker参数中传入colors向量，向量长度要和你的数据行数一致（这里是13个）：

plot_ly(
  dtd7 %>% mutate(n = as.numeric(as.character(n))),
  labels = ~topic,
  parents = NA,
  values = ~n,
  type = 'treemap',
  hovertemplate = "Category: %{label}<br>Percent: %{value}%<extra></extra>",
  # 添加颜色配置
  marker = list(
    colors = c("#FF6B6B", "#4ECDC4", "#45B7D1", "#96CEB4", "#FFEAA7",
               "#DDA0DD", "#98D8C8", "#F7DC6F", "#BB8FCE", "#85C1E9",
               "#F8C471", "#82E0AA", "#F1948A")
  )
)

2. 根据数值自动渐变上色 #

如果想让颜色随n值的大小变化，用color参数绑定数值列，再通过colorscale选择配色方案（Plotly内置了很多配色，比如"Viridis"、"Blues"、"Reds"等）：

plot_ly(
  dtd7 %>% mutate(n = as.numeric(as.character(n))),
  labels = ~topic,
  parents = NA,
  values = ~n,
  type = 'treemap',
  hovertemplate = "Category: %{label}<br>Percent: %{value}%<extra></extra>",
  # 根据n值自动上色
  color = ~n,
  colorscale = "Viridis",
  # 可选：显示颜色条
  colorbar = list(title = "Percent")
)

3. 手动指定每个类别的颜色（关联数据列） #

先在数据框里新增一列存储对应类别的颜色，再绑定到marker的colors参数，方便精准控制：

# 先给数据框添加颜色列
dtd7_colored <- dtd7 %>%
  mutate(
    n = as.numeric(as.character(n)),
    # 给每个topic指定对应颜色
    category_color = case_when(
      topic == "Alcoholic Beverages" ~ "#FF5733",
      topic == "Apparel" ~ "#33FF57",
      topic == "Used Cars and Trucks" ~ "#3357FF",
      # 其他类别依次补充
      TRUE ~ "#CCCCCC" # 默认颜色
    )
  )

# 绘制树形图
plot_ly(
  dtd7_colored,
  labels = ~topic,
  parents = NA,
  values = ~n,
  type = 'treemap',
  hovertemplate = "Category: %{label}<br>Percent: %{value}%<extra></extra>",
  marker = list(colors = ~category_color)
)

内容的提问来源于stack exchange，提问作者Data and AI Nerd