如何在R Markdown中为指定数据框的各列最小值单元格设置加粗格式

我来帮你搞定这个需求！你想要给合并后数据框里df1的每一列最小值所在单元格单独加粗，而不是整行加粗对吧？下面是具体的实现方案，分两种方式供你选择：

方法一：用循环逐个处理单元格 #

这种方式直观易懂，适合新手理解：

set.seed(1)
df1 <- data.frame(
  seed1 = rnorm(5, mean = 0, sd = 1),
  seed2 = rnorm(5, mean = 0, sd = 1),
  seed3 = rnorm(5, mean = 0, sd = 1)
)
set.seed(1)
df2 <- data.frame(
  seed1 = rnorm(5, mean = 0, sd = 2),
  seed2 = rnorm(5, mean = 0, sd = 2),
  seed3 = rnorm(5, mean = 0, sd = 2)
)
df <- cbind(df1, df2)

# 第一步：找到df1每一列最小值对应的行号
min_rows_df1 <- sapply(df1, which.min)
# df1在合并后的df里是前3列，记录列索引
df1_cols <- 1:ncol(df1)

# 第二步：给目标单元格添加加粗格式
df_formatted <- df
for (i in seq_along(df1_cols)) {
  col_idx <- df1_cols[i]
  row_idx <- min_rows_df1[i]
  # 用cell_spec把数值转成带加粗样式的HTML文本
  df_formatted[row_idx, col_idx] <- kableExtra::cell_spec(df[row_idx, col_idx], bold = TRUE, format = "html")
}

# 第三步：生成表格，注意必须设置escape=FALSE！
df_formatted |>
  knitr::kable(format = "html", table.attr = "style='width:100%;'", digits = 2, align = 'c',
               caption = "5 Random Numbers from the Normal Dist at diff set and std dev",
               escape = FALSE) |>  # 禁用转义才能让HTML格式生效
  kableExtra::kable_styling(bootstrap_options = 'bordered') |>
  kableExtra::add_header_above(c('$sd = 1$' = 3, '$sd$ = 2' = 3)) |>
  kableExtra::row_spec(3, bold=T, hline_after = TRUE)

方法二：用dplyr的tidy风格处理（更简洁） #

如果你习惯用tidyverse语法，这种方式更优雅：

library(dplyr)
library(kableExtra)

set.seed(1)
df1 <- data.frame(
  seed1 = rnorm(5, mean = 0, sd = 1),
  seed2 = rnorm(5, mean = 0, sd = 1),
  seed3 = rnorm(5, mean = 0, sd = 1)
)
set.seed(1)
df2 <- data.frame(
  seed1 = rnorm(5, mean = 0, sd = 2),
  seed2 = rnorm(5, mean = 0, sd = 2),
  seed3 = rnorm(5, mean = 0, sd = 2)
)
df <- cbind(df1, df2)

# 直接用mutate+across批量处理df1的列
df_formatted <- df |>
  mutate(across(all_of(names(df1)), function(col) {
    min_val <- min(col)
    # 对等于最小值的单元格设置加粗，其他保持原样
    ifelse(col == min_val, cell_spec(col, bold = TRUE, format = "html"), as.character(col))
  }))

# 生成表格，同样要设置escape=FALSE
df_formatted |>
  knitr::kable(format = "html", table.attr = "style='width:100%;'", digits = 2, align = 'c',
               caption = "5 Random Numbers from the Normal Dist at diff set and std dev",
               escape = FALSE) |>
  kable_styling(bootstrap_options = 'bordered') |>
  add_header_above(c('$sd = 1$' = 3, '$sd$ = 2' = 3)) |>
  row_spec(3, bold=T, hline_after = TRUE)

关键注意点 #

1. escape=FALSE必须设置：如果不关掉转义，kable会把cell_spec生成的HTML标签当成纯文本显示，看不到加粗效果；
2. 只针对df1处理：两种方法都只对df1的列生效，df2的列完全不受影响；
3. 保留了你原来的第三行加粗和下划线设置，不需要的话可以去掉最后一行row_spec代码。

运行上面的代码后，你就能看到df1每一列的最小值单元格被单独加粗，其他内容保持原样啦！

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