Provides a broad catalogue of original categorical, sequential, and diverging colour palettes for scientific graphics, together with palette functions, named discrete and continuous colour and fill scales, catalogue plots, and palette diagnostics for 'ggplot2'. Palettes can also be retrieved as hexadecimal colour vectors for use in base graphics and other visualization systems.
ggpalettes provides 26 original colour palettes and native colour/fill scales
for ggplot2. Its compact pal_*() and scale_*() interface will feel
familiar to users of scientific palette packages such as ggsci, while the
catalogue focuses on balanced, publication-ready colours for the
ggcraft ecosystem.

# install.packages("pak")
pak::pak("YaoxiangLi/ggpalettes")
library(ggpalettes)
library(ggplot2)
gg_palette_names()
gg_palette_info(type = "sequential")
# Palette functions: pal_name()(n)
pal_meadow()(6)
pal_aurora()(12)
ggplot(iris, aes(Sepal.Length, Sepal.Width, colour = Species)) +
geom_point(size = 3) +
scale_colour_clinical() +
theme_minimal()
ggplot(faithfuld, aes(waiting, eruptions, fill = density)) +
geom_raster() +
scale_fill_ember() +
theme_minimal()
Every palette has pal_*(), scale_color_*(), scale_colour_*(), and
scale_fill_*() forms. Generic interfaces are also available:
scale_colour_ggpalette("atelier")
scale_fill_ggpalette("balance")
scale_fill_ggpalette("ocean", binned = TRUE)
Categorical palettes default to discrete scales. Sequential and diverging palettes default to continuous scales, so most plots need no scale-mode argument.
meadow, atelier, clinical, spectrum, pastel, earth,
midnight, floral, coastal, and harvest.aurora, ocean, ember, forest, orchid, slate,
sunrise, glacier, berry, and sand.balance, temperature, earthsky, rosepine, lagoon, and
violetgold.Use gg_palette_plot() to draw all or part of the catalogue. Use
gg_palette_check() to screen CIE Lab separation and contrast against a
selected background.
gg_palette_plot(type = "diverging", n = 9)
gg_palette_check("meadow", background = "white")
The public API is tested on Windows, macOS, and Linux through R CMD check.
MIT © Yaoxiang Li