Question

我有一个散点图，但是由于某种原因，并不是我的所有点都沿着中心对齐。而是，这些点似乎在每个类别的中心网格线上方“浮动”。如何调整他们的位置？

当前图形代码：

ggplot(data = filter(lastDet,
                     !ID %in% c("NB2015.341", "NB2018.7", 
                                "NB2018.43", "NB2018.45")),
       aes(x = speciesEN, y = yday(depDate))) + 
  geom_dotplot(aes(fill = factor(year), colour = factor(year)), 
               binaxis = 'y', stackdir = 'center', dotsize = 0.7,
               stackgroups = TRUE, 
               binpositions = "all", 
               method = "histodot") +
  stat_summary(fun.y = mean, geom = "point",
               colour = "black", shape = 4, size = 4) +
  scale_y_continuous(limits = c(181, max(yday(lastDet$depDate))),
                     breaks = c(181, 196, 212, 227, 243),
                     labels = c("1 Jul", "15 Jul", "1 Aug",
                                "15 Aug", "1 Sep")) + 
  scale_colour_manual(values = c("white", "white", "white"),
                      guide = FALSE) +
  scale_fill_manual(values = c("#CA3542", "#FEBC38", "#849FAD")) + 
  coord_flip() +
  labs(y = "Departure date", 
       x = "", 
       fill = "Year") +
  theme_bw() +
  theme(panel.grid.minor = element_blank())

编辑：

重复出现的问题中提供的“ hacky”解决方案不适用于较大的数据框。

编辑2：

根据要求，这是我数据框的子集。我试图使其更具代表性。

dput(lastDet.sub)
structure(list(speciesEN = c("Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Bank Swallow", "Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Bank Swallow", "Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Bank Swallow", "Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Bank Swallow", "Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Bank Swallow", "Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Bank Swallow", "Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Bank Swallow", "Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Bank Swallow", "Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Bank Swallow", "Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Bank Swallow", "Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Bank Swallow", "Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Bank Swallow", "Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Bank Swallow", "Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Bank Swallow", "Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Bank Swallow", "Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Bank Swallow", "Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Bank Swallow", "Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Bank Swallow", "Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Bank Swallow", "Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Bank Swallow", "Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Bank Swallow", "Bank Swallow", "Bank Swallow", 
"Bank Swallow", "Barn Swallow", "Barn Swallow", "Barn Swallow", 
"Barn Swallow", "Barn Swallow", "Barn Swallow", "Barn Swallow", 
"Barn Swallow", "Barn Swallow", "Barn Swallow", "Barn Swallow", 
"Barn Swallow", "Barn Swallow", "Barn Swallow", "Barn Swallow", 
"Barn Swallow", "Barn Swallow", "Barn Swallow", "Barn Swallow", 
"Barn Swallow", "Cliff Swallow", "Cliff Swallow", "Cliff Swallow", 
"Cliff Swallow", "Cliff Swallow", "Cliff Swallow", "Cliff Swallow", 
"Cliff Swallow", "Cliff Swallow", "Cliff Swallow", "Tree Swallow", 
"Tree Swallow", "Tree Swallow", "Tree Swallow", "Tree Swallow", 
"Tree Swallow", "Tree Swallow", "Tree Swallow", "Tree Swallow", 
"Tree Swallow"), year = c(2018, 2018, 2018, 2018, 2018, 2018, 
2018, 2018, 2018, 2018, 2014, 2014, 2014, 2014, 2014, 2014, 2014, 
2014, 2014, 2014, 2014, 2014, 2014, 2014, 2014, 2014, 2014, 2014, 
2014, 2014, 2014, 2014, 2014, 2014, 2015, 2015, 2015, 2015, 2015, 
2015, 2015, 2015, 2015, 2015, 2015, 2015, 2015, 2015, 2015, 2015, 
2015, 2015, 2015, 2015, 2015, 2015, 2015, 2015, 2015, 2015, 2015, 
2015, 2015, 2015, 2015, 2015, 2015, 2015, 2015, 2015, 2015, 2015, 
2015, 2015, 2015, 2015, 2015, 2015, 2018, 2018, 2018, 2018, 2018, 
2018, 2018, 2018, 2018, 2018, 2018, 2018, 2018, 2018, 2018, 2018, 
2018, 2018, 2018, 2018, 2018, 2018, 2018, 2018, 2018, 2018, 2018, 
2018, 2018, 2018, 2018, 2018, 2018, 2018, 2018, 2018, 2018, 2018, 
2018, 2018, 2018, 2018, 2018, 2018, 2018, 2018, 2018, 2018, 2018
), depDate = structure(c(1531737145.3837, 1534093849.7991, 1530997725.9412, 
1531076069.5422, 1533041446.3001, 1531076076.231, 1533032604.2326, 
1534086023.11, 1531737149.7107, 1532882823.5637, 1406899801.4133, 
1406297348.1112, 1407276094.4158, 1405278154.7126, 1406330632.0613, 
1407490305.4993, 1406370738.3239, 1406655731.0996, 1406552025.8514, 
1406366734.892, 1405371083.9787, 1405935159.2899, 1406656817.3111, 
1405964739.2527, 1405935269.2718, 1407405240.0903, 1406416984.3628, 
1407078925.0925, 1406201767.0139, 1406285074.475, 1407362335.7969, 
1405886217.5548, 1406586545.9097, 1406123078.5834, 1439577132.3188, 
1439594526.7378, 1438097832.985, 1438097743.3089, 1438470481.0184, 
1438246572.18, 1438252727.6824, 1436190169.4567, 1438595642.3985, 
1438728341.4292, 1437298553.0341, 1437573409.4331, 1439402462.271, 
1437577437.0105, 1438209133.1853, 1438006270.5975, 1438424815.9509, 
1438260465.2748, 1439561952.8794, 1441068339.0791, 1437322842.1042, 
1438794640.5246, 1439673494.5053, 1439825890.7982, 1438173337.2322, 
1438805137.1318, 1438083434.1402, 1436372532.3594, 1435960025.2834, 
1439508786.8144, 1439668992.4789, 1439545490.7975, 1438962415.0014, 
1437768970.3123, 1438423416.3447, 1438251288.8178, 1439757591.1317, 
1437844819.4679, 1438247033.9429, 1439141690.47, 1438207292.6573, 
1438171703.2655, 1438277212.7584, 1439199695.1607, 1534507807.6186, 
1532702235.5266, 1533820579.7317, 1532957266.8756, 1531824345.5634, 
1532598380.2699, 1532098828.52, 1533568517.8915, 1533407138.0839, 
1535885104.009, 1533305719.4551, 1535042919.6962, 1533555209.5058, 
1535132109.2126, 1534526297.7543, 1535543329.7974, 1535797758.1752, 
1535538134.3428, 1535279656.0938, 1532184581.5469, 1534778864.0948, 
1534708212.9832, 1534862676.3394, 1535027042.7154, 1534936249.3542, 
1535211828.7767, 1534936232.9197, 1535227479.6166, 1534506480.411, 
1533161753.2014, 1533115007.6647, 1531570748.0769, 1533673583.6144, 
1533201700.8021, 1531239978.3688, 1531843656.4454, 1532604093.761, 
1530790685.643, 1531132089.8288, 1530533431.1395, 1530787169.1414, 
1530471991.9423, 1530643624.7047, 1530474331.461, 1531679979.9343, 
1531673724.9375, 1531579988.1816, 1531394352.5095, 1531831107.0235
), class = c("POSIXct", "POSIXt"), tzone = "UTC")), class = c("tbl_df", 
"tbl", "data.frame"), row.names = c(NA, -127L))

Answer 1

我不经常使用geom_dotplot，所以我不能声称自己对此最熟悉。但是基于样本数据，似乎可以对x进行分面，并将每个x值分配给不同的分面面板，这可以解决对齐问题。看看这是否适用于您的大型数据框？

# original plot as per question, other than the data portion since lastDet has no ID to filter
p1 <- ggplot(data = lastDet,
            aes(x = speciesEN, y = yday(depDate))) + 
  geom_dotplot(aes(fill = factor(year), colour = factor(year)), 
               binaxis = 'y', stackdir = 'center', dotsize = 0.7,
               stackgroups = TRUE, 
               binpositions = "all", 
               method = "histodot") +
  stat_summary(fun.y = mean, geom = "point",
               colour = "black", shape = 4, size = 4) +
  scale_y_continuous(limits = c(181, max(yday(lastDet$depDate))),
                     breaks = c(181, 196, 212, 227, 243),
                     labels = c("1 Jul", "15 Jul", "1 Aug",
                                "15 Aug", "1 Sep")) + 
  scale_colour_manual(values = c("white", "white", "white"),
                      guide = FALSE) +
  scale_fill_manual(values = c("#CA3542", "#FEBC38", "#849FAD")) + 
  coord_flip() +
  labs(y = "Departure date", 
       x = "", 
       fill = "Year") +
  theme_bw() +
  theme(panel.grid.minor = element_blank())

# alignment fixed by facet
p2 <- p1 + 
  facet_grid(forcats::fct_rev(speciesEN) ~ .,     # use fct_rev to reverse factor levels & match
                                                  # x-axis order (can replace with simply
                                                  # `speciesEN ~ .`, if this isn't needed)
             scales = "free", space = "free") +   
  theme(strip.background = element_blank(),
        strip.text = element_blank(),
        panel.spacing = unit(0, "pt"))

默认情况下，每个构面面板的大小均相同。同一位置的点数不会影响比例限制，因此在space = "free"中设置facet_grid不会更改分配给每个构面的空间量。由于“银行燕子”刻面面板的出发日期相同，因此点数要多于其他点，因此我在p2中使用了非常宽的高宽比，以便为每个刻面留出足够的空间。

但是，如果我们负担不起如此大的面板尺寸，我们还可以手动计算最大点数：

rel.width <- layer_data(p2) %>%
  select(PANEL, y, fill, colour, count) %>% 
  unique() %>%
  group_by(PANEL, y) %>%
  summarise(count = sum(count)) %>%
  summarise(count = max(count),
            y = mean(y, na.rm = TRUE)) %>%
  rename(speciesEN = PANEL)
levels(rel.width$speciesEN) <- rev(sort(unique(lastDet$speciesEN)))

> rel.width
# A tibble: 4 x 3
  speciesEN     count     y
  <fct>         <dbl> <dbl>
1 Tree Swallow      2  192.
2 Cliff Swallow     3  200 
3 Barn Swallow      5  229.
4 Bank Swallow     12  210.

并将其添加为绘图中的不可见图层，以强制每个构面调整其空间：

p3 <- p2 +
  geom_tile(data = rel.width,
            aes(y = y, width = count, height = 1),
            alpha = 0)

geom_dotplot中未对齐的点

1 个答案: