我可以使用 geom_text 或类似方法按总和添加组而不是修改原始数据框吗？答案

【问题标题】：Can I use geom_text or similar to add group by sum rather than amending original data frame?我可以使用 geom_text 或类似方法按总和添加组而不是修改原始数据框吗？
【发布时间】：2017-06-22 06:19:50
【问题描述】：

这是一个条形图：

ggplot(filtered_funnel, aes(x = reorder(Funnel, -Sessions), y = Sessions)) +
      geom_bar(stat = "identity", fill = "#008080", alpha = 0.6) +
      xlab("Step") +
      ylab("Events") +
      scale_y_continuous(labels = function(l) {l = l / 1000; paste0(l, "K")}) +
      geom_text(aes(label = Sessions, group = Channel), color = "white")

看起来像这样（注意白色的文本标签）：

这是因为数据 df filtered_funnel 实际上是由字段“Channel”分割出来的。我需要酒吧基于例如的分组总和。会话而不是每个单独的频道会话。

这里是源数据的一瞥：

> glimpse(filtered_funnel)
Observations: 108
Variables: 4
$ Channel  <chr> "Direct", "Direct", "Direct", "Direct", "Direct", "Direct", "Direct", "Direct", "Direct", "Direct", "Direct", "Direct", ...
$ Promo    <chr> "apples", "apples", "apples", "banannas", "banannas", "banannas", "carrots", "carrots", "carrots", "none", "none", "none...
$ Funnel   <chr> "Checkout", "ShippingDetails", "Transactions", "Checkout", "ShippingDetails", "Transactions", "Checkout", "ShippingDetai...
$ Sessions <dbl> 3993, 6332, 2224, 1237, 1962, 689, 2234, 3543, 1244, 42378, 4672, 28120, 87187, 7408, 2602, 611, 969, 340, 4462, 7280, 2...


filtered_funnel
Source: local data frame [108 x 4]
Groups: Channel, Promo [?]

   Channel    Promo          Funnel Sessions
     <chr>    <chr>           <chr>    <dbl>
1   Direct   apples        Checkout     3993
2   Direct   apples ShippingDetails     6332
3   Direct   apples    Transactions     2224
4   Direct banannas        Checkout     1237
5   Direct banannas ShippingDetails     1962
6   Direct banannas    Transactions      689
7   Direct  carrots        Checkout     2234
8   Direct  carrots ShippingDetails     3543
9   Direct  carrots    Transactions     1244
10  Direct     none       AddToCart    42378
# ... with 98 more rows

看起来 ggplot 正在添加每个组件的单个值而不是总和，例如仅用于图像中的第一个栏（会话步骤）

> filtered_funnel %>% filter(Funnel == "Sessions")
Source: local data frame [6 x 4]
Groups: Channel, Promo [6]

   Channel Promo   Funnel Sessions
     <chr> <chr>    <chr>    <dbl>
1   Direct  none Sessions    87187
2    Email  none Sessions   110035
3 Facebook  none Sessions    79734
4  Organic  none Sessions    80768
5      SEM  none Sessions    94610
6  Youtube  none Sessions    66681

我可以在图像和表格中看到值 110035。我真正想要 ggplot 做的是添加会话的总和。

由于这是在 Shiny 应用程序中完成的，因此我试图避免干预源数据，因为我使用过滤器框输入来过滤数据框。此外，我看到其他 SO 答案似乎表明我想做什么是可能的，我只是无法让它工作，例如

draw the sum value above the stacked bar in ggplot2

如何让 ggplot 添加每个条的总和？我可以使用 ggplot 分组和求和而不是更改我在 aes 中输入 ggplot 的源数据吗？

---- 数据输入--- 发表评论后，这是数据（无论如何都是随机生成的，因此没有 NDA 问题）

> dput(filtered_funnel)
structure(list(Channel = c("Direct", "Direct", "Direct", "Direct", 
"Direct", "Direct", "Direct", "Direct", "Direct", "Direct", "Direct", 
"Direct", "Direct", "Direct", "Direct", "Direct", "Direct", "Direct", 
"Email", "Email", "Email", "Email", "Email", "Email", "Email", 
"Email", "Email", "Email", "Email", "Email", "Email", "Email", 
"Email", "Email", "Email", "Email", "Facebook", "Facebook", "Facebook", 
"Facebook", "Facebook", "Facebook", "Facebook", "Facebook", "Facebook", 
"Facebook", "Facebook", "Facebook", "Facebook", "Facebook", "Facebook", 
"Facebook", "Facebook", "Facebook", "Organic", "Organic", "Organic", 
"Organic", "Organic", "Organic", "Organic", "Organic", "Organic", 
"Organic", "Organic", "Organic", "Organic", "Organic", "Organic", 
"Organic", "Organic", "Organic", "SEM", "SEM", "SEM", "SEM", 
"SEM", "SEM", "SEM", "SEM", "SEM", "SEM", "SEM", "SEM", "SEM", 
"SEM", "SEM", "SEM", "SEM", "SEM", "Youtube", "Youtube", "Youtube", 
"Youtube", "Youtube", "Youtube", "Youtube", "Youtube", "Youtube", 
"Youtube", "Youtube", "Youtube", "Youtube", "Youtube", "Youtube", 
"Youtube", "Youtube", "Youtube"), Promo = c("apples", "apples", 
"apples", "banannas", "banannas", "banannas", "carrots", "carrots", 
"carrots", "none", "none", "none", "none", "none", "none", "pears", 
"pears", "pears", "apples", "apples", "apples", "banannas", "banannas", 
"banannas", "carrots", "carrots", "carrots", "none", "none", 
"none", "none", "none", "none", "pears", "pears", "pears", "apples", 
"apples", "apples", "banannas", "banannas", "banannas", "carrots", 
"carrots", "carrots", "none", "none", "none", "none", "none", 
"none", "pears", "pears", "pears", "apples", "apples", "apples", 
"banannas", "banannas", "banannas", "carrots", "carrots", "carrots", 
"none", "none", "none", "none", "none", "none", "pears", "pears", 
"pears", "apples", "apples", "apples", "banannas", "banannas", 
"banannas", "carrots", "carrots", "carrots", "none", "none", 
"none", "none", "none", "none", "pears", "pears", "pears", "apples", 
"apples", "apples", "banannas", "banannas", "banannas", "carrots", 
"carrots", "carrots", "none", "none", "none", "none", "none", 
"none", "pears", "pears", "pears"), Funnel = c("Checkout", "ShippingDetails", 
"Transactions", "Checkout", "ShippingDetails", "Transactions", 
"Checkout", "ShippingDetails", "Transactions", "AddToCart", "Checkout", 
"Registrations", "Sessions", "ShippingDetails", "Transactions", 
"Checkout", "ShippingDetails", "Transactions", "Checkout", "ShippingDetails", 
"Transactions", "Checkout", "ShippingDetails", "Transactions", 
"Checkout", "ShippingDetails", "Transactions", "AddToCart", "Checkout", 
"Registrations", "Sessions", "ShippingDetails", "Transactions", 
"Checkout", "ShippingDetails", "Transactions", "Checkout", "ShippingDetails", 
"Transactions", "Checkout", "ShippingDetails", "Transactions", 
"Checkout", "ShippingDetails", "Transactions", "AddToCart", "Checkout", 
"Registrations", "Sessions", "ShippingDetails", "Transactions", 
"Checkout", "ShippingDetails", "Transactions", "Checkout", "ShippingDetails", 
"Transactions", "Checkout", "ShippingDetails", "Transactions", 
"Checkout", "ShippingDetails", "Transactions", "AddToCart", "Checkout", 
"Registrations", "Sessions", "ShippingDetails", "Transactions", 
"Checkout", "ShippingDetails", "Transactions", "Checkout", "ShippingDetails", 
"Transactions", "Checkout", "ShippingDetails", "Transactions", 
"Checkout", "ShippingDetails", "Transactions", "AddToCart", "Checkout", 
"Registrations", "Sessions", "ShippingDetails", "Transactions", 
"Checkout", "ShippingDetails", "Transactions", "Checkout", "ShippingDetails", 
"Transactions", "Checkout", "ShippingDetails", "Transactions", 
"Checkout", "ShippingDetails", "Transactions", "AddToCart", "Checkout", 
"Registrations", "Sessions", "ShippingDetails", "Transactions", 
"Checkout", "ShippingDetails", "Transactions"), Sessions = c(3993, 
6332, 2224, 1237, 1962, 689, 2234, 3543, 1244, 42378, 4672, 28120, 
87187, 7408, 2602, 611, 969, 340, 4462, 7280, 2304, 549, 896, 
283, 2094, 3417, 1081, 42251, 5666, 29094, 110035, 9244, 2926, 
256, 418, 132, 129, 191, 85, 3078, 4557, 2039, 120, 178, 79, 
13977, 90, 9727, 79734, 134, 59, 1142, 1691, 756, 3125, 4655, 
1985, 1724, 2568, 1095, 3109, 4631, 1975, 34756, 2864, 23453, 
80768, 4266, 1819, 249, 371, 158, 1839, 2661, 1223, 1543, 2232, 
1026, 2007, 2904, 1335, 24090, 1792, 15272, 94610, 2593, 1192, 
479, 693, 318, 800, 1245, 522, 1734, 2698, 1132, 930, 1447, 607, 
22349, 1436, 14478, 66681, 2235, 937, 1579, 2457, 1031)), class = c("grouped_df", 
"tbl_df", "tbl", "data.frame"), row.names = c(NA, -108L), drop = TRUE, .Names = c("Channel", "Promo", 
"Funnel", "Sessions"), indices = list(0:2, 3:5, 6:8, 9:14, 15:17, 
    18:20, 21:23, 24:26, 27:32, 33:35, 36:38, 39:41, 42:44, 45:50, 
    51:53, 54:56, 57:59, 60:62, 63:68, 69:71, 72:74, 75:77, 78:80, 
    81:86, 87:89, 90:92, 93:95, 96:98, 99:104, 105:107), group_sizes = c(3L, 
3L, 3L, 6L, 3L, 3L, 3L, 3L, 6L, 3L, 3L, 3L, 3L, 6L, 3L, 3L, 3L, 
3L, 6L, 3L, 3L, 3L, 3L, 6L, 3L, 3L, 3L, 3L, 6L, 3L), biggest_group_size = 6L, labels = structure(list(
    Channel = c("Direct", "Direct", "Direct", "Direct", "Direct", 
    "Email", "Email", "Email", "Email", "Email", "Facebook", 
    "Facebook", "Facebook", "Facebook", "Facebook", "Organic", 
    "Organic", "Organic", "Organic", "Organic", "SEM", "SEM", 
    "SEM", "SEM", "SEM", "Youtube", "Youtube", "Youtube", "Youtube", 
    "Youtube"), Promo = c("apples", "banannas", "carrots", "none", 
    "pears", "apples", "banannas", "carrots", "none", "pears", 
    "apples", "banannas", "carrots", "none", "pears", "apples", 
    "banannas", "carrots", "none", "pears", "apples", "banannas", 
    "carrots", "none", "pears", "apples", "banannas", "carrots", 
    "none", "pears")), class = "data.frame", row.names = c(NA, 
-30L), drop = TRUE, .Names = c("Channel", 
"Promo")))

【问题讨论】：

使用dput(filtered_funnel)分享您的数据
您好现在添加了这个，感谢您的帮助

标签： r ggplot2

【解决方案1】：

您可以使用 stat_summary 计算总和，并将其用于 y 位置和标签：

ggplot(filtered_funnel, aes(x = reorder(Funnel, -Sessions), y = Sessions)) +
  geom_bar(stat = "identity", fill = "#008080", alpha = 0.6) +
  stat_summary(
    aes(label = stat(y)), fun.y = 'sum', geom = 'text', col = 'white', vjust = 1.5
  ) +
  scale_y_continuous(labels = function(l) {l = l / 1000; paste0(l, "K")})  +
  xlab("Step") + ylab("Events")

【讨论】：

非常感谢（再次）！这很棒。我可以问一些后续吗？什么是..y..？你为什么选择（你怎么知道选择）stat_summary() 而不是geom_text()？
..y.. 是计算的统计量。每个统计数据（identity 除外）都会在绘图之前计算一些值，您可以通过用.. 包围的名称访问这些值。您可以通过stats或geoms定义层，通常没关系，您也可以使用geom_text(aes(label = ..y..), fun.y = 'sum', stat = 'summary', col = 'white', vjust = 1.5)。
谢谢，这真的很有帮助。 “..y.. 是计算的统计量”。计算出来的统计数据是什么？ fun.y = "sum"?我还能在标签上添加什么（甚至是一些荒谬的东西），让它真正点击？（希望您不介意后续跟进，我真的很重视您在有机会时提供的帮助）
stat_summary 为每个x 汇总y。所以它计算一个新的y。您可以使用..y.. 访问这个新的y。如果您使用..x..，您会得到数字 1 到 6。对于其他统计数据，您可以访问其他计算值。例如，如果您查看docs page 中的stat_bin，则有一个“计算变量”部分，其中包含多个选项，例如ndensity。当使用stat_bin 作为stat（直接或在geom 中）时，您可以使用..ndensity.. 访问此变量。
"stat_summary 为每个 x 总结 y"。点击。非常感谢！