Topolearnggplot2 is the standard for clear, publication-quality graphics in R.
ggplot2 implements the grammar of graphics: a plot is data plus mappings from columns to visual properties, drawn by geometric layers.
ggplot(df, aes(x = wt, y = mpg))sets the data and default aesthetics (x,y,colour,fill,size,shape).- Add layers with
+:geom_point(),geom_line(),geom_col()(bar heights from data),geom_bar()(counts),geom_histogram(),geom_boxplot(),geom_smooth(method = "lm"). facet_wrap(~ cyl)splits into small multiples.labs(title =, x =, y =)sets labels,scale_*functions control axes and colours, andtheme_minimal()and friends change the look.
Mapping versus setting is the classic confusion: aes(colour = factor(cyl)) maps a column to colour and adds a legend, while geom_point(colour = "blue") outside aes sets a constant.
ggplot2 wants long data: one row per observation, one column per variable. If you find yourself adding one geom_line per series, reshape with tidyr::pivot_longer and map the series name to colour instead.
Save with ggsave("plot.png", p, width = 6, height = 4, dpi = 300). Inside scripts and functions, a plot only appears when printed, so call print(p).
Example
library(ggplot2)
p <- ggplot(mtcars, aes(x = wt, y = mpg, colour = factor(cyl))) +
geom_point(size = 2) +
geom_smooth(method = "lm", se = FALSE, formula = y ~ x) +
facet_wrap(~ am, labeller = label_both) +
labs(title = "Fuel economy vs weight", x = "Weight (1000 lb)",
y = "Miles per gallon", colour = "Cylinders") +
theme_minimal()
out <- tempfile(fileext = ".png")
ggsave(out, p, width = 7, height = 4, dpi = 100)
print(file.exists(out)) #> [1] TRUE
print(length(p$layers)) #> [1] 2Remember
- A plot = data +
aes()mappings + geom layers added with+ - Inside
aes()maps a column; outside sets a constant - Reshape to long data instead of one layer per series
ggsavewrites files;print(p)shows plots inside scripts
Watch out for
- Writing
aes(colour = "blue"), which maps the constant string as a category and adds a legend.