rework entire text for latest cowplot and dviz.supp packages.

Author: clauswilke

_bookdown.yml

@@ -1,4 +1,6 @@
 new_session: yes
+output_dir: "docs"
+delete_merged_file: false
 
 rmd_files: [
   "index.Rmd",
@@ -10,15 +12,15 @@ rmd_files: [
   "colorblind.Rmd",
   "redundant_coding.Rmd",
   "small_axis_labels.Rmd",
-  #"aspect_ratio.Rmd",
+  ##"aspect_ratio.Rmd",
   "background_grids.Rmd",
   "avoid_line_drawings.Rmd",
-  "proportional_ink.Rmd",
+  ##"proportional_ink.Rmd",
   "overlapping_points.Rmd",
 
   # Part II
   "directory_of_visualizations.Rmd",
-  #"visualizing_amounts.Rmd",
+  ##"visualizing_amounts.Rmd",
   "visualizing_proportions.Rmd",
   "visualizing_distributions_I.Rmd",
   "visualizing_distributions_II.Rmd",
@@ -31,5 +33,3 @@ rmd_files: [
 
   "references.Rmd"
 ]
-
-output_dir: "docs"

_build.sh

@@ -0,0 +1,3 @@
+#!/bin/sh
+
+Rscript -e "bookdown::render_book('index.Rmd', 'bookdown::gitbook')"

_common.R

@@ -3,6 +3,8 @@ options(digits = 3)
 
 knitr::opts_chunk$set(
   echo = FALSE,
+  message = FALSE,
+  warnings = FALSE,
   cache = FALSE,
   fig.align = 'center',
   fig.width = 6,
@@ -12,7 +14,4 @@ knitr::opts_chunk$set(
 
 options(dplyr.print_min = 6, dplyr.print_max = 6)
 
-library(cowplot)
-
-source("R/misc.R")
-source("R/themes.R")
+library(dviz.supp)

avoid_line_drawings.Rmd

@@ -29,7 +29,7 @@ h3_bad <- ggplot(age_hist_3, aes(x = age, y = count)) + geom_col(width = 2., fil
                                                              color = "black")  + 
   scale_y_continuous(limits = c(0, 86), expand = c(0, 0), breaks = 25*(0:5)) +
   scale_x_continuous(limits = c(0, 75), expand = c(0, 0)) +
-  theme_dviz() +
+  theme_half_open() +
   background_grid(major = "y", minor = "none")
 
 stamp_bad(h3_bad)
@@ -41,7 +41,7 @@ stamp_bad(h3_bad)
 h3_good <- ggplot(age_hist_3, aes(x = age, y = count)) + geom_col(width = 2.7, fill = "#56B4E9")  + 
   scale_y_continuous(limits = c(0, 86), expand = c(0, 0), breaks = 25*(0:5)) +
   scale_x_continuous(limits = c(0, 75), expand = c(0, 0)) +
-  theme_dviz() +
+  theme_half_open() +
   background_grid(major = "y", minor = "none")
 
 stamp_good(h3_good)
@@ -70,13 +70,14 @@ iris_lines <- ggplot(iris_dens, aes(x = Sepal.Length, y = density, linetype = Sp
   geom_density_line(stat = "identity", alpha = 0., size = 0.75) +
   geom_text(data = iris_max, aes(label = Species, hjust = hjust, vjust = vjust,
                                  x = Sepal.Length + nudge_x, 
-                                 y = density + nudge_y), size = 12/.pt) +
+                                 y = density + nudge_y),
+            inherit.aes = FALSE, size = 12/.pt) +
   scale_linetype_manual(values = c(1, 3, 4),
                     breaks = c("virginica", "versicolor", "setosa"),
                     guide = "none") +
   scale_x_continuous(expand = c(0, 0), name = "sepal length") +
   scale_y_continuous(limits = c(0, 1.5), expand = c(0, 0)) +
-  theme_dviz()
+  theme_half_open()
 
 stamp_bad(iris_lines)
 ```
@@ -90,13 +91,14 @@ iris_colored_lines <- ggplot(iris_dens, aes(x = Sepal.Length, y = density, color
   geom_density_line(stat = "identity", alpha = 0., size = 1) +
   geom_text(data = iris_max, aes(label = Species, hjust = hjust, vjust = vjust,
                                  x = Sepal.Length + nudge_x, 
-                                 y = density + nudge_y), size = 12/.pt) +
+                                 y = density + nudge_y),
+            inherit.aes = FALSE, size = 12/.pt) +
   scale_color_manual(values = c("#56B4E9", "#E69F00", "#009E73"),
                     breaks = c("virginica", "versicolor", "setosa"),
                     guide = "none") +
   scale_x_continuous(expand = c(0, 0), name = "sepal length") +
   scale_y_continuous(limits = c(0, 1.5), expand = c(0, 0)) +
-  theme_dviz()
+  theme_half_open()
 
 stamp_ugly(iris_colored_lines)
 ```
@@ -109,13 +111,14 @@ iris_filled <- ggplot(iris_dens, aes(x = Sepal.Length, y = density, fill = Speci
   geom_density_line(stat = "identity", alpha = 0.7) +
   geom_text(data = iris_max, aes(label = Species, hjust = hjust, vjust = vjust,
                                  x = Sepal.Length + nudge_x, 
-                                 y = density + nudge_y), size = 12/.pt) +
+                                 y = density + nudge_y),
+            inherit.aes = FALSE, size = 12/.pt) +
   scale_fill_manual(values = c("#56B4E9", "#E69F00", "#009E73"),
                     breaks = c("virginica", "versicolor", "setosa"),
                     guide = "none") +
   scale_x_continuous(expand = c(0, 0), name = "sepal length") +
   scale_y_continuous(limits = c(0, 1.5), expand = c(0, 0)) +
-  theme_dviz()
+  theme_half_open()
   
 stamp_good(iris_filled)
 ```

background_grids.Rmd

@@ -65,7 +65,7 @@ We could try to remove the grid altogether, but I think that is a worse option (
 (ref:price-plot-no-grid) Indexed stock price over time for four major tech companies.
 
 ```{r price-plot-no-grid, warning=FALSE, fig.cap='(ref:price-plot-no-grid)'}
-stamp_bad(price_plot + theme_dviz_half_open())
+stamp_bad(price_plot + theme_half_open())
 ```
 
 (ref:price-plot-refline) Indexed stock price over time for four major tech companies.
@@ -84,9 +84,7 @@ price_plot2 <- ggplot(prices, aes(x=date, y=price_indexed, color=ticker)) +
   scale_y_continuous(name="stock price, indexed",
                      limits = c(0, 560),
                      expand=c(0,0)) +
-  theme_dviz_half_open()
-
-
+  theme_half_open()
 
 stamp_good(price_plot2)
 ```
@@ -96,7 +94,7 @@ Alternatively, we can use just a minimal grid. In particular, for a plot where w
 (ref:price-plot-hgrid) Indexed stock price over time for four major tech companies.
 
 ```{r price-plot-hgrid, warning=FALSE, fig.cap='(ref:price-plot-hgrid)'}
-stamp_good(price_plot + theme_dviz_hgrid())
+stamp_good(price_plot + theme_minimal_hgrid())
 ```
 
 For such a minimal grid, we generally draw the lines orthogonally to direction along which the numbers of interest vary. Therefore, if instead of plotting the stock price over time we plot the five-year increase, as horizontal bars, then we will want to use vertical lines instead (Figure \@ref(fig:price-increase)).
@@ -122,7 +120,7 @@ perc_plot <- ggplot(perc_increase, aes(x=ticker, y=perc)) +
                      breaks=c("FB", "GOOG", "MSFT", "AAPL"),
                      labels=c("Facebook", "Alphabet", "Microsoft", "Apple")) +
   coord_flip() +
-  theme_dviz_vgrid() +
+  theme_minimal_vgrid() +
   theme(axis.title.y=element_blank()) # remove the unnecessary space generated by an empty label
 
 stamp_good(perc_plot)
@@ -150,7 +148,7 @@ iris %>%
     scale_x_continuous(limits=c(1.95, 4.6), expand=c(0, 0)) +
     scale_y_continuous(limits=c(3.95, 8.1), expand=c(0, 0)) -> iris_plot
 
-stamp_good(plot_grid(NULL, iris_plot + theme_dviz_grid(), NULL, nrow=1, rel_widths=c(0.02, 1, 0.04)))
+stamp_good(plot_grid(NULL, iris_plot + theme_minimal_grid(), NULL, nrow=1, rel_widths=c(0.02, 1, 0.04)))
 ```
 
 For figures where the relevant comparison is the *x* = *y* line, I prefer to draw a diagonal line rather than a grid. For example, consider Figure \@ref(fig:echave-et-al), which compares two sets of correlations for 209 protein structures. By drawing the diagonal line, we can see immediately which correlations are systematically stronger. The same observation is much harder to make when the figure has a background grid instead (Figure \@ref(fig:echave-et-al-grid)). Thus, even though Figure \@ref(fig:echave-et-al-grid) looks pleasing, I label it as bad.
@@ -167,7 +165,8 @@ nrg_plot <- ggplot(cor_table, aes(y=cor.RSA, x=-cor.WCN)) +
   coord_fixed() +
   xlab("rate–WCN correlation") +
   ylab("rate–RSA correlation") +
-  labs(title = "") # add an empty title as a simple trick to move the plot downwards
+  labs(title = "") +
+  theme_half_open() # add an empty title as a simple trick to move the plot downwards
 
 stamp_good(plot_grid(nrg_plot, NULL, nrow=1, rel_widths=c(1, 0.2)))
 ```
@@ -184,8 +183,9 @@ nrg_plot2 <- ggplot(cor_table, aes(y=cor.RSA, x=-cor.WCN)) +
   coord_fixed() +
   xlab("rate–WCN correlation") +
   ylab("rate–RSA correlation") +
-  background_grid(size.major=0.5) +
   labs(title = "") # add an empty title as a simple trick to move the plot downwards
+  theme_half_open() +
+  background_grid(size.major=0.5)
 
 stamp_bad(plot_grid(nrg_plot2, NULL, nrow=1, rel_widths=c(1, 0.2)))
 ```

boxplots_violins.Rmd

@@ -6,7 +6,6 @@ library(dplyr)
 library(forcats)
 library(ggridges)
 library(ggforce)
-library(viridis)
 ```
 
 # Boxplots, violin plots, and more {#boxplots-violins}
@@ -36,7 +35,8 @@ lincoln_errbar <- ggplot(lincoln_df, aes(x = month_short, y = `Mean Temperature
   stat_summary(fun.y = mean, fun.ymax = function(x) {mean(x) + 2*sd(x)},
                fun.ymin = function(x) {mean(x) - 2*sd(x)}, geom = "pointrange",
                fatten = 5) +
-  xlab("Month")
+  xlab("Month") + 
+  theme_half_open()
 
 stamp_bad(lincoln_errbar)
 ```
@@ -63,8 +63,8 @@ The line in the middle represents the median, and the box encloses the middle 50
 When we visualize the temperature dataset using boxplots, we obtain the following result.
 ```{r boxplots, message=FALSE, warning=FALSE}
 lincoln_box <- ggplot(lincoln_df, aes(x = month_short, y = `Mean Temperature [F]`)) +
-  geom_boxplot(fill = 'grey90') + xlab("Month")
-
+  geom_boxplot(fill = 'grey90') + xlab("Month") +
+  theme_half_open()
 
 stamp_good(lincoln_box)
 ```
@@ -74,7 +74,8 @@ Using the boxplot visualization, we see clearly that temperature is highly skewe
 
 ```{r violin-plots, message=FALSE, warning=FALSE}
 lincoln_violin <- ggplot(lincoln_df, aes(x = month_short, y = `Mean Temperature [F]`)) +
-  geom_violin(fill = 'grey90') + xlab("Month")
+  geom_violin(fill = 'grey90') + xlab("Month") +
+  theme_half_open()
 
 stamp_good(lincoln_violin)
 ```
@@ -83,15 +84,17 @@ stamp_good(lincoln_violin)
 We can also plot all individual points:
 ```{r all-points-no-jitter, message=FALSE, warning=FALSE}
 lincoln_points <- ggplot(lincoln_df, aes(x = month_short, y = `Mean Temperature [F]`)) +
-  geom_point() + xlab("Month")
+  geom_point() + xlab("Month") +
+  theme_half_open()
 
 stamp_bad(lincoln_points)
 ```
 
 
 ```{r jittered-points, message=FALSE, warning=FALSE}
 lincoln_jitter <- ggplot(lincoln_df, aes(x = month_short, y = `Mean Temperature [F]`)) +
-  geom_jitter(width = .15) + xlab("Month")
+  geom_jitter(width = .15) + xlab("Month") +
+  theme_half_open()
 
 stamp_good(lincoln_jitter)
 ```
@@ -100,7 +103,10 @@ stamp_good(lincoln_jitter)
 Finally, we can combine the best of both worlds and spread the dots out in proportion to the number of points with a similar y coordinate. This methods yields the sina plot, which shows each individual dot while also visualizing the distributions. 
 ```{r sina-plots, message=FALSE, warning=FALSE}
 lincoln_sina <- ggplot(lincoln_df, aes(x = month_short, y = `Mean Temperature [F]`)) +
-  geom_violin(color = "transparent", fill = "gray90") + stat_sina(maxwidth = 0.8) + xlab("Month")
+  geom_violin(color = "transparent", fill = "gray90") + 
+  stat_sina(maxwidth = 0.8) + 
+  xlab("Month") +
+  theme_half_open()
 
 stamp_good(lincoln_sina)
 ```

colorblind.Rmd

@@ -107,7 +107,7 @@ You can see examples using these scales throughout this book, for example in the
 
 
 
-```{r} 
+```{r eval = FALSE, include = FALSE} 
 ## Notes, entered as R comment so as to not appear in the printed output
 #Some useful posts on color-blind friendly palettes:
 #

datasets/Australian_athletes/scatter_plot.R

@@ -1,6 +1,7 @@
 library(DAAG)
 library(ggplot2)
 library(ggridges)
+library(dplyr)
 
 male_ais <- filter(ais, sex=="m") %>%
   filter(sport %in% c("B_Ball", "Field", "Swim", "T_400m", "T_Sprnt", "W_Polo")) %>%
@@ -14,9 +15,9 @@ male_ais <- filter(ais, sex=="m") %>%
 
 male_ais$sport <- factor(male_ais$sport, levels = c("field", "water polo", "basketball", "swim", "track"))
 
-ggplot(male_ais, aes(x=ht, y=pcBfat, color=sport, fill=sport, shape=sport)) + 
-  geom_point(size = 2, alpha = 0.7) +
+ggplot(male_ais, aes(x=ht, y=pcBfat, color=sport, fill=sport, shape=sport)) +
+  geom_point(size = 3) +
   scale_shape_manual(values = 21:25) +
   scale_color_OkabeIto(order=c(2, 1, 3, 4, 5), darken = 0.3) +
-  scale_fill_OkabeIto(order=c(2, 1, 3, 4, 5), darken = 0.1) +
+  scale_fill_OkabeIto(order=c(2, 1, 3, 4, 5), darken = 0.1, alpha = 0.5) +
   theme_minimal()

directory_of_visualizations.Rmd

@@ -5,7 +5,6 @@ source("_common.R")
 library(dplyr)
 library(ggforce)
 library(ggridges)
-library(colorspace)
 ```
 
 # (PART\*) Part II: A visualization for every occasion {-}
@@ -19,7 +18,7 @@ library(colorspace)
 theme_plot_icon <- function(bg_color = "#F5F8EA", line_color = "#243400") {
   line_size <- .8
   font_size <- 10
-  theme_dviz_half_open() %+replace% theme(
+  theme_half_open() %+replace% theme(
     axis.text.x       = element_blank(),
     axis.text.y       = element_blank(),
     axis.title.x      = element_blank(),