r - y轴ggplot上未显示log10 y轴标签； 无法将地层标签推到垂直中心(log10 y-axis labels do not appear on R ggaluvial ggplot ; cannot nudge strata labels to vertical center)

Question

A compound question:

(一个复合问题：)

• I cannot get log10 y-axis labels to usefully display on the ggalluvial plot after many variations of scale_y_log10.

  (经过很多scale_y_log10的变化之后，我无法获得log10 y轴标签有效地显示在积积图上。)

  I am having particular difficulty specifying and formatting breaks.

  (我在指定和格式化中断时遇到特别困难。)

  scale_y_continuous produces the following:

  (scale_y_continuous产生以下内容：)

Error in seq.default(min, max, by = by) : 'to' must be a finite number`

• Strata labels obscure the strata boundaries.

  (地层标签遮盖了地层边界。)

  The geom_label nudge_y parameter has no apparent effect.

  (geom_label nudge_y参数没有明显作用。)

  How should one center these labels in the strata?

  (如何将这些标签放在地层中心？)

  The default setting do not accomplish this.

  (默认设置不能完成此操作。)

Please see plot, code and data below.

(请参阅下面的图，代码和数据。)

Appreciate any advice.

(感谢任何建议。)

library(tidyverse)
library(ggplot2)
library(ggalluvial)
library(scales)
#library(ggthemes) # for theme_economist

proAntiByActivity = ggplot(as.data.frame(a),
       aes(y = aggSpend,
           axis1 = cycle, axis3 = proAnti, axis2 = activityGroup)) +
  #geom_alluvium(aes(fill =cycle))+ #, width = 0, knot.pos = 0, reverse = FALSE) +
  geom_alluvium(aes(fill =cycle), width = 1/12, knot.pos = 1/6, reverse = FALSE, show.legend = TRUE) +
  guides(fill = FALSE) +
  geom_stratum(width = 1/5, reverse = FALSE) +
  #geom_text(stat = "stratum", label.strata = TRUE, reverse = FALSE) +
  geom_label(stat = "stratum", label.strata = TRUE,nudge_y = 0) +
  #scale_x_discrete(breaks = 1:3, labels = c("Election Cycle","Means",  "Pro/Anti")) +
  scale_x_continuous(breaks = 1:3, labels = c("Election Cycle","Means",  "Pro/Anti")) +
  #scale_y_continuous(name="total spent")+
  #scale_y_continuous(trans = "log10",name="total spent",limits=NULL) +
  #coord_trans(y="log10")+
  #scale_y_log10(name="log total spent",breaks = 1e+100*c(2e+03,2e+04,2e+05,2e+06,2e+07,2e+08,2e+09), labels = c(2e+03,2e+04,2e+05,2e+06,2e+07,2e+08,2e+09)) +
  scale_y_log10(name="log total spent",breaks = trans_breaks("log10", function(x) 10^x), labels = trans_format("log10", scientific_format())) +
  #scale_y_log10(name="log total spent",breaks = trans_breaks("log10", function(x) 10^x), labels = trans_format("log10", math_format(.x))) +
#scale_y_log10(name="log total spent",breaks = trans_breaks("log10", function(x) 10^x), labels = trans_format("log10", math_format(10^.x))) +
  #theme_economist()+
  ggtitle("Money Flows for U.S. Elections By Activity 2010-2020")

proAntiByActivity

Plot data

(绘图数据)

a<-structure(list(cycle = c("2010", "2010", "2010", "2010", "2010", 
"2010", "2010", "2010", "2010", "2010", "2010", "2010", "2010", 
"2010", "2010", "2010", "2010", "2010", "2010", "2010", "2010", 
"2010", "2010", "2010", "2010", "2010", "2010", "2010", "2010", 
"2010", "2010", "2010", "2010", "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"), proAnti = c("antiDem", 
"antiDem", "antiDem", "antiDem", "antiDem", "antiDem", "antiGOP", 
"antiGOP", "antiGOP", "antiGOP", "antiGOP", "antiGOP", "antiOther", 
"antiOther", "antiOther", "antiOther", "antiOther", "proDem", 
"proDem", "proDem", "proDem", "proDem", "proDem", "proGOP", "proGOP", 
"proGOP", "proGOP", "proGOP", "proGOP", "proOther", "proOther", 
"proOther", "proOther", "antiDem", "antiDem", "antiDem", "antiDem", 
"antiDem", "antiDem", "antiGOP", "antiGOP", "antiGOP", "antiGOP", 
"antiGOP", "antiGOP", "antiOther", "antiOther", "antiOther", 
"antiOther", "antiOther", "proDem", "proDem", "proDem", "proDem", 
"proDem", "proDem", "proGOP", "proGOP", "proGOP", "proGOP", "proGOP", 
"proGOP", "proOther", "proOther", "proOther", "proOther", "proOther", 
"proOther"), activityGroup = c("Advertizing", "AdvertizingV2", 
"Outreach", "OutreachV2", "Overhead", "Strategy", "Advertizing", 
"AdvertizingV2", "Outreach", "OutreachV2", "Overhead", "Strategy", 
"Advertizing", "AdvertizingV2", "Outreach", "Overhead", "Strategy", 
"Advertizing", "AdvertizingV2", "Outreach", "OutreachV2", "Overhead", 
"Strategy", "Advertizing", "AdvertizingV2", "Outreach", "OutreachV2", 
"Overhead", "Strategy", "Advertizing", "Outreach", "Overhead", 
"Strategy", "Advertizing", "AdvertizingV2", "Outreach", "OutreachV2", 
"Overhead", "Strategy", "Advertizing", "AdvertizingV2", "Outreach", 
"OutreachV2", "Overhead", "Strategy", "Advertizing", "AdvertizingV2", 
"Outreach", "Overhead", "Strategy", "Advertizing", "AdvertizingV2", 
"Outreach", "OutreachV2", "Overhead", "Strategy", "Advertizing", 
"AdvertizingV2", "Outreach", "OutreachV2", "Overhead", "Strategy", 
"Advertizing", "AdvertizingV2", "Outreach", "OutreachV2", "Overhead", 
"Strategy"), aggSpend = c(159948962.660721, 40399.6402031971, 
9158355.18213395, 283702.113259935, 187198.211058457, 4563675.63724594, 
152982928.325557, 216256.874555558, 7993712.18735522, 823.580861836789, 
106873.732663197, 1778558.91739064, 1438915.09226517, 146.251315353075, 
1028161.44634096, 25685.7644535242, 196881.375812725, 23407657.0206254, 
133629.176224183, 14368371.0220397, 26389.4225797726, 1539957.71484811, 
1538921.52699396, 21854214.1121801, 470029.494152242, 11779580.9624892, 
522106.091640242, 791874.639134986, 2898174.32990864, 8.14755986834644, 
11010.7374694267, 5596.20969242995, 3061.26236033165, 311475306.900278, 
4535009.26475469, 14866419.8988148, 46627.7825986496, 139697.610043845, 
5685839.23231515, 442793202.646583, 9185627.87610291, 25475752.6745331, 
17801.571227516, 854976.211853977, 4386843.04877027, 16497196.406344, 
164121.870068956, 831667.837824925, 3.90373318422706, 129470.611578985, 
118114661.837676, 5903094.13420614, 43261891.7968911, 369371.87531218, 
2141258.36101987, 4361747.5737002, 69866481.4457689, 4214523.71305267, 
25400537.1038862, 242901.276117105, 92174.6149800344, 14024469.650107, 
6396957.39009888, 200539.496487483, 736056.134792953, 833.390308236151, 
2446.50392841341, 285246.096898001)), class = c("grouped_df", 
"tbl_df", "tbl", "data.frame"), row.names = c(NA, -68L), groups = structure(list(
    cycle = c("2010", "2010", "2010", "2010", "2010", "2010", 
    "2018", "2018", "2018", "2018", "2018", "2018"), proAnti = c("antiDem", 
    "antiGOP", "antiOther", "proDem", "proGOP", "proOther", "antiDem", 
    "antiGOP", "antiOther", "proDem", "proGOP", "proOther"), 
    .rows = list(1:6, 7:12, 13:17, 18:23, 24:29, 30:33, 34:39, 
        40:45, 46:50, 51:56, 57:62, 63:68)), row.names = c(NA, 
-12L), class = c("tbl_df", "tbl", "data.frame"), .drop = TRUE))

  ask by Steve Dutky translate from so

Answer 1

The following is an alternative to ggalluvial using ggforce.

(以下是使用ggforce进行ggalluvial的替代方法。)

I find that the spacing ggforce places between strata helps the understanding of the plot by resolving categories.

(我发现各层之间的距离ggforce位置通过解决类别有助于理解绘图。)

Assume a is the dput output you've posted.

(假设a是您发布的dput输出。)

library(tidyverse)
library(ggforce)

df <- gather_set_data(a, 1:3)

ggplot(df, aes(x = x, id = id, split = y, value = log10(aggSpend))) +
  geom_parallel_sets(aes(fill = as.factor(cycle)), alpha = 0.5,
                     axis.width = 0.5) +
  geom_parallel_sets_axes(axis.width = 0.5, fill = "white", colour = "black") +
  geom_parallel_sets_labels(angle = 0) +
  scale_x_discrete(limits = c("cycle", "activityGroup", "proAnti"),
                   labels = c("Election Cycle", "Means", "Pro/Anti"),
                   name = "") +
  scale_y_continuous(name = "Log10 Total Spending") +
  guides(fill = "none")

With regard to the log10 axes, my recommendation is to title the y-axis appropriately and do a log10 transform on the data itself rather than the axis.

(关于log10轴，我的建议是适当地命名y轴，并对数据本身而不是轴进行log10转换。)

See my comment underneath your question too for my objections.

(对于我的异议，也请参阅您问题下方的我的评论。)