Package {ggthemes}

ggthemes: Extra Themes, Scales and Geoms for 'ggplot2'

Description

Some extra themes, geoms, and scales for 'ggplot2'. Provides 'ggplot2' themes and scales that replicate the look of plots by Edward Tufte, Stephen Few, 'Fivethirtyeight', 'The Economist', 'Stata', 'Excel', and 'The Wall Street Journal', among others. Provides 'geoms' for Tufte's box plot and range frame.

Author(s)

Maintainer: Jeffrey B. Arnold jeffrey.arnold@gmail.com (<https://orcid.org/0000-0001-9953-3904>)

Authors:

• Jeffrey B. Arnold jeffrey.arnold@gmail.com (<https://orcid.org/0000-0001-9953-3904>)

Other contributors:

• Gergely Daroczi [contributor]

• Bo Werth [contributor]

• Brian Weitzner [contributor]

• Joshua Kunst [contributor]

• Baptiste Auguie [contributor]

• Bob Rudis [contributor]

• Hadley Wickham (Code from the ggplot2 package.) [contributor]

• Justin Talbot (Code from the labeling package) [contributor]

• Joshua London [contributor]

See Also

Useful links:

• https://jrnold.github.io/ggthemes/

• https://github.com/jrnold/ggthemes

• Report bugs at https://github.com/jrnold/ggthemes/issues

Bank a Plot's Own Data to 45 Degrees

Description

A convenience wrapper around bank_slopes that extracts x/y directly from an already-specified ggplot, so you do not have to separately reconstruct the plotted vectors by hand. It builds plot with ggplot_build, computes the banking ratio from one layer's fully resolved data (i.e. after stats, position adjustments, and faceting have been applied), and returns plot + coord_fixed(ratio = ...).

Usage

bank_plot(
  plot,
  method = c("ms", "as", "ao", "was"),
  cull = FALSE,
  layer = 1,
  ...
)

Arguments

Details

Segments are never averaged across a group or facet panel boundary: slopes are computed within each combination of group and PANEL and then combined, so a line plot with multiple series (or facets) is banked correctly rather than picking up spurious slopes between the end of one line and the start of the next.

Note that coord_fixed applies a single ratio to every panel, so faceted plots are banked using the combined data from all panels rather than a ratio tailored to each one individually.

Value

The plot, with coord_fixed added.

See Also

bank_slopes

Examples

library("ggplot2")
x <- seq_along(sunspot.year)
y <- as.numeric(sunspot.year)
p <- ggplot(data.frame(x = x, y = y), aes(x = x, y = y)) +
  geom_line()
bank_plot(p)

Bank a Plot's Own Data at Every Scale

Description

A convenience wrapper around bank_slopes_multiscale that extracts y directly from an already-specified ggplot and returns one copy of the plot per scale of interest, each with the appropriate coord_fixed applied. The result is the small-multiples display used throughout Heer and Agrawala (2006): the same data, banked to reveal trends at different frequencies.

Usage

bank_plot_multiscale(
  plot,
  method = c("ms", "as", "ao", "was"),
  cull = TRUE,
  layer = 1,
  ...
)

Arguments

Details

Multi-scale banking is defined on the frequency domain of a single series sampled on a regular grid, so unlike bank_plot this function requires the chosen layer to hold exactly one series with evenly spaced x values.

Value

A named list of ggplot objects, one per retained scale, in ascending order of frequency and named by frequency index.

References

Heer, Jeffrey and Maneesh Agrawala, 2006. "Multi-Scale Banking to 45." IEEE Transactions On Visualization And Computer Graphics 12(5).

See Also

bank_slopes_multiscale, bank_plot

Examples

library("ggplot2")

y <- as.numeric(sunspot.year)
p <- ggplot(data.frame(x = seq_along(y), y = y), aes(x = x, y = y)) +
  geom_line()

# One plot per scale of interest, named by frequency index.
plots <- bank_plot_multiscale(p)
names(plots)

## Low-frequency trend across sunspot cycles
plots[[1]]

## The individual 11-year cycles
plots[[2]]

Bank Slopes to 45 degrees

Description

Calculate the optimal aspect ratio of a line graph by banking the slopes to 45 degrees as suggested by W.S. Cleveland. This maximizes the ability to visually differentiate differences in slope. This function will calculate the optimal aspect ratio for a line plot using any of the methods described in Heer and Agrawala (2006). In their review of the methods they suggest using median absolute slope banking ('ms'), which produces aspect ratios which are generally the median of the various methods provided here.

Usage

bank_slopes(
  x,
  y,
  cull = FALSE,
  weight = NULL,
  method = c("ms", "as", "ao", "was"),
  ...
)

Arguments

Value

numeric The aspect ratio (x , y).

Methods

As written, all of these methods calculate the aspect ratio (x /y), but bank_slopes will return (y / x) to be compatible with link[ggplot2]{coord_fixed()}.

Median Absolute Slopes Banking

Let the aspect ratio be \alpha = \frac{w}{h} then the median absolute slop banking is the \alpha such that,

median \left| \frac{s_i}{\alpha} \right| = 1

Let R_z = z_{max} - z_{min} for z = x, y, and M = median \| s_i \|. Then,

\alpha = M \frac{R_x}{R_y}

Average Absolute Slope Banking

Let the aspect ratio be \alpha = \frac{w}{h}. then the mean absolute slope banking is the \alpha such that,

mean \left| \frac{s_i}{\alpha} \right| = 1

Average Absolute Orientation Banking

Rather than averaging the slopes themselves, this method averages the orientation (angle) of each segment, since perceived slope differences are more closely related to angle than to the raw ratio dy/dx. Let s'_i = s_i R_x / R_y be the range-normalized slopes. Then \alpha is chosen such that,

mean \left| \arctan \left( \frac{s'_i}{\alpha} \right) \right| = \frac{\pi}{4}

This has no closed-form solution and is found numerically with uniroot.

Weighted Average Absolute Orientation Banking

This is the weighted version of Average Absolute Orientation Banking from Heer and Agrawala (2006). Each segment's absolute orientation is weighted by its length in display space, so both the orientation and its weight depend on \alpha. With s'_i as above and segment run dx_i, \alpha is chosen such that,

\frac{\sum_i \left|\arctan(s'_i / \alpha)\right| dx_i \sqrt{1 + (s'_i / \alpha)^2}} {\sum_i dx_i \sqrt{1 + (s'_i / \alpha)^2}} = \frac{\pi}{4}

This has no closed-form solution and is found numerically with uniroot.

All of these methods consider the entirety of the data at once, so they accentuate local features and can obscure larger-scale trends. Heer and Agrawala (2006) address this with multi-scale banking, which uses spectral analysis to identify the frequency scales present in the data and banks each one separately; see bank_slopes_multiscale and bank_plot_multiscale.

References

Cleveland, W. S., M. E. McGill, and R. McGill. The Shape Parameter of a Two-Variable Graph. Journal of the American Statistical Association, 83:289-300, 1988

Heer, Jeffrey and Maneesh Agrawala, 2006. 'Multi-Scale Banking to 45' IEEE Transactions On Visualization And Computer Graphics.

Cleveland, W. S. 1993. 'A Model for Studying Display Methods of Statistical Graphs.' Journal of Computational and Statistical Graphics.

Cleveland, W. S. 1994. The Elements of Graphing Data, Revised Edition.

See Also

banking(), bank_plot to bank a ggplot using its own data, and bank_slopes_multiscale to bank each frequency scale in the data separately.

Examples

library("ggplot2")

# Use the classic sunspot data from Cleveland's original paper
x <- seq_along(sunspot.year)
y <- as.numeric(sunspot.year)
# Without banking
m <- ggplot(data.frame(x = x, y = y), aes(x = x, y = y)) +
  geom_line()
m

## Using the default method, Median Absolute Slope
ratio <- bank_slopes(x, y)
m + coord_fixed(ratio = ratio)

## Average Absolute Slope
m + coord_fixed(ratio = bank_slopes(x, y, method = "as"))

## Average Absolute Orientation
m + coord_fixed(ratio = bank_slopes(x, y, method = "ao"))

## Weighted Average Absolute Slope: each segment is weighted by its run in
## x, so this only differs from "as" when x is not evenly spaced
m + coord_fixed(ratio = bank_slopes(x, y, method = "was"))

## Culling removes slopes of 0 or Inf before banking, which matters when
## the data contains runs of repeated x or y values
bank_slopes(x, y, cull = TRUE)

Multi-Scale Banking to 45 Degrees

Description

Compute a set of aspect ratios, one per frequency scale present in a series, using the multi-scale banking algorithm of Heer and Agrawala (2006). Single-scale banking (bank_slopes) considers the whole series at once, so it accentuates local features and can obscure larger-scale trends. Multi-scale banking instead uses spectral analysis to find the scales that carry real energy, low-pass filters the data to each of those scales in turn, and banks the resulting trend curve, yielding one aspect ratio per scale.

Usage

bank_slopes_multiscale(
  y,
  method = c("ms", "as", "ao", "was"),
  cull = TRUE,
  window = 3,
  sd = 1,
  threshold = NULL,
  scale_factor = 1.25
)

Arguments

Details

The procedure is Algorithm 1 of Heer and Agrawala (2006):

1. Take the discrete Fourier transform of y and form the power spectrum from the squared coefficient magnitudes.

2. Smooth the spectrum by convolving it with a Gaussian kernel, since spectral energy tends to arrive in "clumps" containing local oscillation.

3. Threshold the smoothed spectrum. Contiguous runs above the threshold are collapsed to their highest-frequency bin, capturing the total contribution of that region of energy.

4. For each retained scale, low-pass filter y to remove all higher frequencies and bank the resulting trend curve to 45 degrees using bank_slopes.

5. Discard aspect ratios within scale_factor of the previous retained ratio, since they would produce visually redundant charts.

The scale corresponding to the data in its entirety is always included.

Because the algorithm is defined on the frequency domain of y alone, it assumes observations are evenly spaced in x; the banking of each trend curve uses x = seq_along(y).

Value

A tibble with one row per retained scale, in ascending order of frequency, and columns:

frequency

integer frequency index, i.e. the number of times the trend repeats across the series.

ratio

numeric aspect ratio in the y / x sense used by coord_fixed().

aspect_ratio

numeric the same value as width / height, the convention in which the banking literature reports aspect ratios.

References

Heer, Jeffrey and Maneesh Agrawala, 2006. "Multi-Scale Banking to 45." IEEE Transactions On Visualization And Computer Graphics 12(5).

Cleveland, W. S. 1993. "A Model for Studying Display Methods of Statistical Graphs." Journal of Computational and Statistical Graphics.

See Also

bank_slopes for single-scale banking, and bank_plot_multiscale to bank a ggplot at every scale.

Examples

library("ggplot2")

# Sunspot activity, the classic example from Cleveland and from Heer and
# Agrawala's Section 3.2.1. Spectral analysis identifies scales at frequency
# indices 7, 10, 31 and 36, plus the data in its entirety; culling similar
# aspect ratios leaves two charts worth drawing.
y <- as.numeric(sunspot.year)
bank_slopes_multiscale(y)

# `ratio` is ready for coord_fixed(); `aspect_ratio` is the same value as
# width / height, the convention used in the banking literature.
scales <- bank_slopes_multiscale(y)

m <- ggplot(data.frame(x = seq_along(y), y = y), aes(x = x, y = y)) +
  geom_line()

## The low-frequency trend: the oscillation of high points across cycles
m + coord_fixed(ratio = scales$ratio[[1]])

## The 11-year cycle: a steep onset followed by a more gradual decay
m + coord_fixed(ratio = scales$ratio[[2]])

## Raise the threshold to select fewer scales
bank_slopes_multiscale(y, threshold = Inf)

## Any of the single-scale banking methods can be used for each scale
bank_slopes_multiscale(y, method = "ao")

Calc color palette (discrete)

Description

Color palettes from LibreOffice Calc. This palette has 12 values.

Usage

calc_pal()

Value

A palette function. It takes the number of colours n and returns a character vector of n hex colours, and can be used as the palette argument of discrete_scale().

See Also

Other colour calc: scale_fill_calc()

Examples

library("scales")

show_col(calc_pal()(12))

Calc shape palette (discrete)

Description

Shape palette based on the shapes used in LibreOffice Calc.

Usage

calc_shape_pal(unicode = FALSE)

Arguments

Value

A palette function. It takes the number of shapes n and returns an integer vector of n shape (pch) codes, and can be used as the palette argument of discrete_scale().

Note

This palette supports seven values by default and thirteen with unicode = TRUE. Six of Calc's thirteen symbols – the solid down, left and right triangles, the bowtie, the hourglass and the four-pointed star – have no base pch equivalent and are dropped rather than approximated by a different shape. Restoring them with unicode = TRUE needs a font covering Geometric Shapes, Dingbats and Miscellaneous Mathematical Symbols-B; Noto Sans Symbols 2 is effectively the only free font with the last of these.

See Also

Other shapes calc: scale_shape_calc()

Examples

# The seven shapes with a font-independent equivalent.
show_shapes(calc_shape_pal()(7))

## Not run: 
  # All thirteen, drawn from the device font. Needs a font covering Geometric
  # Shapes, Dingbats and Miscellaneous Mathematical Symbols-B, such as
  # Noto Sans Symbols 2.
  show_shapes(calc_shape_pal(unicode = TRUE)(13))

## End(Not run)

Canva.com color palettes

Description

150+ color palettes from canva.com. See canva_palettes().

Usage

canva_pal(palette = "Fresh and bright")

Arguments

Value

A function that takes a single value, the number of colors to use.

Examples

require("ggplot2")
require("tibble")
if (require("purrr") && require("scales") && require("dplyr")) {
  canva_df <- map2_df(
    canva_palettes,
    names(canva_palettes),
    ~ tibble(
      colors = .x,
      .id = seq_along(colors),
      palette = .y
    )
  )
  ggplot(
    canva_df,
    aes(
      y = palette,
      x = .id,
      fill = colors
    )
  ) +
    geom_raster() +
    scale_fill_identity(guide = FALSE) +
    theme_minimal() +
    theme(
      panel.grid = element_blank(),
      axis.text.x = element_blank()
    ) +
    labs(x = "", y = "")

  show_col(canva_pal("Fresh and bright")(4))
  show_col(canva_pal("Cool blues")(4))
  show_col(canva_pal("Modern and crisp")(4))
}

150 Color Palettes from Canva

Description

150 four-color palettes by the canva.com design school. These palettes were derived from photos and "impactful websites".

Usage

canva_palettes

Format

A named list of character vector. The names are the palette names. The values of the character vectors are hex colors, e.g. "#f98866".

References

• Janie Kliever, 100 Brilliant Color Combinations and How to Apply Them to Your Designs, Canva.com, June 20, 2015.

• Mary Stribley, Website Color Schemes: The Palettes of 50 Visually Impactful Websites to Inspire You, Canva.com, January 26, 2016.

• Schwabish, Jonathan. 150+ Color Palettes for Excel, PolicyViz, January 12, 2017.

Examples

require("ggplot2")
require("tibble")
if (require("purrr") && require("scales") && require("dplyr")) {
  canva_df <- map2_df(
    canva_palettes,
    names(canva_palettes),
    ~ tibble(
      colors = .x,
      .id = seq_along(colors),
      palette = .y
    )
  )
  ggplot(
    canva_df,
    aes(
      y = palette,
      x = .id,
      fill = colors
    )
  ) +
    geom_raster() +
    scale_fill_identity(guide = FALSE) +
    theme_minimal() +
    theme(
      panel.grid = element_blank(),
      axis.text.x = element_blank()
    ) +
    labs(x = "", y = "")

  show_col(canva_pal("Fresh and bright")(4))
  show_col(canva_pal("Cool blues")(4))
  show_col(canva_pal("Modern and crisp")(4))
}

Filled Circle Shape palette (discrete)

Description

'r lifecycle::badge("deprecated")'

This function was deprecated because unicode glyphs used for the circles vary in size, making them unusable for plotting.

Shape palette with circles varying by amount of fill. This uses the set of 3 circle fill values in Lewandowsky and Spence (1989): solid, hollow, half-filled, with two additional fill amounts: three-quarters, and one-quarter.

This palette supports up to five values.

Usage

circlefill_shape_pal()

Value

A palette function. It takes the number of shapes n and returns an integer vector of n shape (pch) codes, and can be used as the palette argument of discrete_scale().

References

Lewandowsky, Stephan and Ian Spence (1989) "Discriminating Strata in Scatterplots", Journal of the American Statistical Association, https://www.jstor.org/stable/2289649

See Also

Other shapes: cleveland_shape_pal(), scale_shape_circlefill(), scale_shape_cleveland(), scale_shape_tremmel(), tremmel_shape_pal()

Examples

circlefill_shape_pal()(3)

Shape palette from Cleveland "Elements of Graphing Data" (discrete).

Description

Shape palettes for overlapping and non-overlapping points.

Usage

cleveland_shape_pal(overlap = TRUE, unicode = FALSE)

Arguments

Value

A palette function. It takes the number of shapes n and returns an integer vector of n shape (pch) codes, and can be used as the palette argument of discrete_scale().

Note

In the Elements of Graphing Data, W.S. Cleveland suggests two shape palettes for scatter plots: one for overlapping data and another for non-overlapping data. The symbols for overlapping data rely on pattern discrimination, while the symbols for non-overlapping data vary the amount of fill.

Following Tremmel (1995), the circle with a vertical line is replaced by an encircled plus sign.

cleveland_shape_pal(overlap = TRUE) supports four values on either branch.

cleveland_shape_pal(overlap = FALSE) supports three values by default and five with unicode = TRUE. Its five source symbols encode fill fraction, which base pch cannot express, so the two partially filled circles are dropped rather than approximated by a different shape. To encode a proportion, map alpha or fill instead; to restore the five glyphs, use unicode = TRUE with a font that covers Mathematical Operators, such as STIX Two Text.

The truncation is arguably an improvement. Tremmel (1995) Experiment 2 tested exactly this five-symbol set and found the fill-graded circles the worst performers measured, with the encircled plus and encircled dot the slowest pair and the one producing the most errors. The three shapes that survive are the better-separating subset.

References

Cleveland WS. The Elements of Graphing Data. Revised Edition. Hobart Press, Summit, NJ, 1994, pp. 154-164, 234-239.

Tremmel, Lothar, (1995) "The Visual Separability of Plotting Symbols in Scatterplots", Journal of Computational and Graphical Statistics, https://www.jstor.org/stable/1390760

See Also

Other shapes: circlefill_shape_pal(), scale_shape_circlefill(), scale_shape_cleveland(), scale_shape_tremmel(), tremmel_shape_pal()

Examples

library("ggplot2")

p <- ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, shape = factor(gear))) +
  facet_wrap(~am) +
  theme_bw()
# overlapping symbol palette
p + scale_shape_cleveland()
# non-overlapping symbol palette
p + scale_shape_cleveland(overlap = FALSE)

Colorblind Color Palette (Discrete) and Scales

Description

An eight-color colorblind safe qualitative discrete palette.

'r lifecycle::badge("deprecated")'

Usage

colorblind_pal(black = TRUE)

colourblind_pal(black = TRUE)

scale_colour_colourblind(black = TRUE, ...)

scale_colour_colorblind(black = TRUE, ...)

scale_color_colorblind(black = TRUE, ...)

scale_fill_colorblind(black = TRUE, ...)

scale_fill_colourblind(black = TRUE, ...)

Arguments

Value

colorblind_pal() and colourblind_pal() return a palette function that takes the number of colours n and returns a character vector of n hex colours. The scale_*() functions return a ggplot2 scale object.

References

Chang, W. "Cookbook for R"

⁠https://jfly.iam.u-tokyo.ac.jp/color⁠

See Also

The dichromat package, dichromat_pal(), and scale_color_tableau() for other colorblind palettes.

Examples

library("ggplot2")
library("scales")

show_col(colorblind_pal()(8))
show_col(colorblind_pal(black = FALSE)(7))
p <- ggplot(mtcars) +
  geom_point(aes(
    x = wt,
    y = mpg,
    colour = factor(gear)
  )) +
  facet_wrap(~am)
p + theme_igray() + scale_colour_colourblind()

Economist color palette (discrete)

Description

The nine colors The Economist uses for chart series: blue, cyan, green, yellow, olive, purple, gold, gray, and red, in that order. Red comes last because the house style reserves it for data the chart is making a point about, rather than handing it out as an ordinary series color.

Usage

economist_pal(fill = deprecated())

Arguments

Details

A tenth color, "Econ red", is the brighter masthead red used for the tag rectangle and for single-series highlights. It is excluded from this palette; take it from ggthemes_data$economist$main when you need it.

Value

A palette function. It takes the number of colours n and returns a character vector of n hex colours, and can be used as the palette argument of discrete_scale().

See Also

Other colour economist: economist_seq_pal(), scale_colour_economist(), scale_colour_economist_c()

Examples

library("scales")

show_col(economist_pal()(6))

## the full set of nine series colours
show_col(economist_pal()(9))

Economist sequential color palettes

Description

The Economist's "equal lightness colour scales": six ordered steps for each of the nine hues in the main chart palette, running darkest to lightest. Use them for ordered data, where the main palette is for unordered categories.

Usage

economist_seq_pal(hue = "blue")

economist_gradient_pal(hue = "blue")

Arguments

Details

economist_seq_pal() returns the six steps themselves, for discrete ordered data. economist_gradient_pal() interpolates between them, for continuous data.

Value

economist_seq_pal() returns a palette function that takes the number of colours n and returns the first n of the six hex colours for hue, for use with discrete_scale(). economist_gradient_pal() returns a palette function that takes a numeric vector x of values between 0 and 1 and returns hex colours interpolated between those steps, for use with continuous_scale().

See Also

Other colour economist: economist_pal(), scale_colour_economist(), scale_colour_economist_c()

Examples

library("scales")

# the six steps of one hue, darkest first
show_col(economist_seq_pal("blue")(6))
show_col(economist_seq_pal("red")(6))

# interpolated for continuous data
show_col(economist_gradient_pal("green")(seq(0, 1, length.out = 10)))

Excel (current versions) color palettes (discrete)

Description

Color palettes used by current versions of Microsoft Office and Excel.

Usage

excel_new_pal(theme = "Office")

Arguments

Details

In 2023 Microsoft replaced the long-standing Office theme with a new default and renamed the old one. The default here, "Office", is the current palette; "Office 2013" is the palette Excel used from 2013 until 2022. The former ggthemes names "Office Theme" and "Office 2007-2010" still work, and select "Office 2013" and "Office 2007" respectively.

Value

A palette function. It takes the number of colours n and returns a character vector of n hex colours, and can be used as the palette argument of discrete_scale().

See Also

Other colour excel: excel_pal(), scale_colour_excel_new(), scale_fill_excel()

Examples

library("scales")

for (i in names(ggthemes::ggthemes_data$excel$themes)) {
  show_col(excel_new_pal(theme = i)(6))
}

Excel 97 ugly color palettes (discrete)

Description

The color palettes used in Microsoft Excel 97 (and up until Excel 2007). Use this for that classic ugly look and feel. For ironic purposes only. 3D bars and pies not included. Please never use this color palette.

Usage

excel_pal(line = TRUE)

Arguments

Value

A palette function. It takes the number of colours n and returns a character vector of n hex colours, and can be used as the palette argument of discrete_scale().

See Also

Other colour excel: excel_new_pal(), scale_colour_excel_new(), scale_fill_excel()

Examples

library("scales")

show_col(excel_pal()(7))
show_col(excel_pal(line = FALSE)(7))

Pretty axis breaks inclusive of extreme values

Description

This function returns pretty axis breaks that always include the extreme values of the data. This works by calling the extended Wilkinson algorithm (Talbot et al., 2010), constrained to solutions interior to the data range. Then, the minimum and maximum labels are moved to the minimum and maximum of the data range.

Usage

extended_range_breaks_(
  dmin,
  dmax,
  n = 5,
  Q = c(1, 5, 2, 2.5, 4, 3),
  w = c(0.25, 0.2, 0.5, 0.05)
)

extended_range_breaks(n = 5, ...)

Arguments

Details

extended_range_breaks_ implements the algorithm and returns the break values. extended_range_breaks uses the conventions of the scales package, and returns a function.

Note that ggplot2 hands a breaks function the expanded scale limits, not the data range. Passing extended_range_breaks() directly to breaks therefore places the end labels at the panel edges rather than at the extremes of the data. To label the extremes, which is what pairs with geom_rangeframe(), apply the function to the data instead: scale_x_continuous(breaks = extended_range_breaks()(mtcars$wt)).

Value

For extended_range_breaks_, the vector of axis label locations. For extended_range_breaks, a function which takes a single argument, a vector of data, and returns the vector of axis label locations.

Author(s)

Justin Talbot jtalbot@stanford.edu, Jeffrey B. Arnold, Baptiste Auguie

References

Talbot, J., Lin, S., Hanrahan, P. (2010) An Extension of Wilkinson's Algorithm for Positioning Tick Labels on Axes, InfoVis 2010.

See Also

Other tufte: geom_rangeframe(), geom_tufteboxplot(), theme_tufte()

Examples

# Pretty breaks that always include the extremes of the data
extended_range_breaks_(min(mtcars$wt), max(mtcars$wt))

# A function, in the style of the scales package
extended_range_breaks()(mtcars$wt)

library("ggplot2")

ggplot(mtcars, aes(x = wt, y = mpg)) +
  geom_point() +
  geom_rangeframe() +
  scale_x_continuous(breaks = extended_range_breaks()(mtcars$wt)) +
  scale_y_continuous(breaks = extended_range_breaks()(mtcars$mpg)) +
  theme_tufte()

Color Palettes Few "Show Me the Numbers"

Description

Qualitative color palettes from Stephen Few (2012) Show Me the Numbers. There are three palettes: Light, Medium, and Dark. Each palette comprises nine colors: gray, blue, orange, green, pink, brown, purple, yellow, red. For n = 1, gray is used. For n > 1, the eight non-gray colors are used.

Usage

few_pal(palette = "Medium")

Arguments

Details

Use the light palette for filled areas, such as bar charts. Use the medium palette for points and lines. Use the dark palette for highlighting specific points or for small and thin lines and points.

Value

A palette function. It takes the number of colours n and returns a character vector of n hex colours, and can be used as the palette argument of discrete_scale().

References

Few, S. (2012) Show Me the Numbers: Designing Tables and Graphs to Enlighten. 2nd edition. Analytics Press.

"Practical Rules for Using Color in Charts".

See Also

Other colour few: scale_colour_few()

Examples

library("scales")

show_col(few_pal()(7))
show_col(few_pal("Dark")(7))
show_col(few_pal("Light")(7))

Shape palette from "Show Me the Numbers" (discrete)

Description

Shape palette from Stephen Few's, "Show Me the Numbers". The shape palette consists of five shapes: circle, square, triangle, plus, times.

Usage

few_shape_pal()

Value

A palette function. It takes the number of shapes n and returns an integer vector of n shape (pch) codes, and can be used as the palette argument of discrete_scale().

References

Few, S. (2012) Show Me the Numbers: Designing Tables and Graphs to Enlighten, Analytics Press, p. 208.

Examples

few_shape_pal()(5)

show_shapes(few_shape_pal()(5))

FiveThirtyEight color palette

Description

The standard three-color FiveThirtyEight palette for line plots comprises blue, red, and green.

Usage

fivethirtyeight_pal()

Value

A palette function. It takes the number of colours n and returns a character vector of n hex colours, and can be used as the palette argument of discrete_scale().

See Also

Other colour fivethirtyeight: scale_colour_fivethirtyeight()

Examples

library("scales")

show_col(fivethirtyeight_pal()(3))

Google Docs color palette (discrete)

Description

Color palettes from Google Docs. This palette includes 20 colors.

Usage

gdocs_pal()

Value

A palette function. It takes the number of colours n and returns a character vector of n hex colours, and can be used as the palette argument of discrete_scale().

See Also

Other colour gdocs: scale_fill_gdocs()

Examples

library("scales")

show_col(gdocs_pal()(24))

Range Frames

Description

Axis lines which extend to the maximum and minimum of the plotted data.

Usage

geom_rangeframe(
  mapping = NULL,
  data = NULL,
  stat = "identity",
  position = "identity",
  ...,
  sides = "bl",
  na.rm = FALSE,
  show.legend = NA,
  inherit.aes = TRUE
)

Arguments

Details

This should be used with 'coord_cartesian(clip="off")' in order to correctly draw the lines.

Secondary axes (sec_axis()) only relabel the existing axis; they do not introduce a separate data range. Because of this, sides = "trbl" already draws correctly-positioned frames on the top/right edges of a panel that has a secondary axis – there is no separate "secondary" range for geom_rangeframe() to draw against.

Value

A ggplot2 layer.

Aesthetics

• colour

• size

• linetype

• alpha

References

Tufte, Edward R. (2001) The Visual Display of Quantitative Information, Chapter 6.

See Also

Other geom tufte: geom_tufteboxplot()

Other tufte: extended_range_breaks_(), geom_tufteboxplot(), theme_tufte()

Examples

library("ggplot2")

ggplot(mtcars, aes(wt, mpg)) +
  geom_point() +
  geom_rangeframe() +
  coord_cartesian(clip = "off") +
  theme_tufte()

# In the example above,
# `coord_cartesian(clip="off")`` ensures that the full width of the line is drawn.
# if you know a better way to fix this,
# please open an issue or PR on github https://github.com/jrnold/ggthemes/issue

# sides = "trbl" also works with a secondary axis: the secondary axis only
# relabels the existing scale, so the frame is still correctly positioned.
ggplot(mtcars, aes(wt, mpg)) +
  geom_point() +
  geom_rangeframe(sides = "trbl") +
  scale_y_continuous(sec.axis = sec_axis(~ . * 0.4251, name = "km/L")) +
  coord_cartesian(clip = "off") +
  theme_tufte()

Tufte's Box Plot

Description

Edward Tufte's revisions of the box plot as described in The Visual Display of Quantitative Information. This functions provides several box plot variants:

• A point indicating the median, a gap indicating the interquartile range, and lines for whiskers.

• An offset line indicating the interquartile range and a gap indicating the median.

• A line indicating the interquartile range, a gap indicating the median, and points indicating the minimum and maximum values

• A wide line indicating the interquartile range, a gap indicating the median, and lines indicating the minimum and maximum.

Usage

geom_tufteboxplot(
  mapping = NULL,
  data = NULL,
  stat = "fivenumber",
  position = "dodge",
  outlier.colour = "black",
  outlier.shape = 19,
  outlier.size = 1.5,
  outlier.stroke = 0.5,
  voffset = 0.01,
  hoffset = 0.005,
  na.rm = FALSE,
  show.legend = NA,
  inherit.aes = TRUE,
  median.type = "point",
  whisker.type = "line",
  ...
)

Arguments

Value

A ggplot2 layer.

Aesthetics

• x [required]

• y [required]

• colour

• size

• linewidth

• linetype

• shape

• fill

• alpha

References

Tufte, Edward R. (2001) The Visual Display of Quantitative Information, Chapter 6.

McGill, R., Tukey, J. W. and Larsen, W. A. (1978) Variations of box plots. The American Statistician 32, 12-16.

See Also

geom_boxplot()

Other geom tufte: geom_rangeframe()

Other tufte: extended_range_breaks_(), geom_rangeframe(), theme_tufte()

Examples

library("ggplot2")

p <- ggplot(mtcars, aes(factor(cyl), mpg))
# with a point for the median and lines for whiskers
p + geom_tufteboxplot()
# with a line for the interquartile range and points for whiskers
p + geom_tufteboxplot(median.type = "line", whisker.type = "point", hoffset = 0)
# with a wide line for the interquartile range and lines for whiskers.
# `width` scales the median line but is also the box width, so use
# position = "identity" to stop position_dodge() warning about overlaps.
p + geom_tufteboxplot(median.type = "line", hoffset = 0, width = 3, position = "identity")
# with an offset line for the interquartile range and lines for whiskers
p + geom_tufteboxplot(median.type = "line")
# combined with theme_tufte
p + geom_tufteboxplot() + theme_tufte() + theme(axis.ticks.x = element_blank())
# traditional boxplot with whiskers only out to 1.5 IQR, outlier points
p + geom_tufteboxplot(stat = "boxplot", outlier.shape = 5)

Palette and theme data

Description

The ggthemes environment contains various values used in themes and palettes. This is undocumented and subject to change.

Usage

ggthemes_data

Format

A list object.

Details

ggthemes_data$stata$colors$names spans both generations of Stata's named colors: the classic set plus the gs0–gs16 gray scale, and the stc1–stc15 colors added in Stata 18. See stata_pal().

Highcharts color palette (discrete)

Description

Highcharts uses many different color palettes in its plots. This collects the palettes shipped with Highcharts 13, plus the default Highcharts used before v11.

Usage

hc_pal(palette = "default")

Arguments

Details

"default" and "default_dark" are the light- and dark-mode forms of the palette Highcharts has used by default since v11.0.0; they differ only in positions 2 and 3. "classic" is the default Highcharts used from v5.0.0 through v10.x. The remaining palettes come from the themes bundled with Highcharts.

Note that "avocado" and "sunset" have only four colors.

Value

A palette function. It takes the number of colours n and returns a character vector of n hex colours, and can be used as the palette argument of discrete_scale().

See Also

Other colour hc: scale_colour_hc()

Examples

hc_pal()(6)

scales::show_col(hc_pal("darkunica")(4))

Apple Numbers color palettes (discrete)

Description

Color palettes used by charts in Apple Numbers. Each palette provides the six series colors that Numbers assigns to a chart's data series, in order.

Usage

numbers_pal(palette = "Classic")

Arguments

Value

A palette function. It takes the number of colours n and returns a character vector of n hex colours, and can be used as the palette argument of discrete_scale().

See Also

Other colour numbers: scale_fill_numbers()

Examples

library("scales")

show_col(numbers_pal()(6))
show_col(numbers_pal("Spectrum")(6))

Color palette from the pander package

Description

The pander ships with a default colorblind and printer-friendly color palette borrowed from ⁠https://jfly.iam.u-tokyo.ac.jp/color/⁠.

Usage

palette_pander(n, random_order = FALSE)

Arguments

Value

A character vector of n hex colours, recycled if n exceeds the number of colours available. Unlike the other *_pal() functions, this is itself the palette function.

See Also

Other colour pander: scale_color_pander()

Examples

palette_pander(8)
# the same colors in a random order
palette_pander(8, random_order = TRUE)

Color Palettes from Paul Tol's "Colour Schemes"

Description

ptol_pal() was deprecated in ggthemes 7.0.0. Use the khroma package instead, which tracks Paul Tol's colour schemes as he revises them.

This palette is the 12-colour qualitative scheme from Tol's 2012 technical note, and has not followed the revisions he has made since. His current schemes are at https://sronpersonalpages.nl/~pault/; the closest successor to this palette is khroma::colour("muted").

Qualitative color palettes from Paul Tol, "Colour Schemes".

Usage

ptol_pal()

Details

Incorporation of the palette into an R package was originally inspired by Peter Carl's Paul Tol 21 Gun Salute

Value

A palette function. It takes the number of colours n and returns a character vector of n hex colours, and can be used as the palette argument of discrete_scale().

References

Paul Tol. 2012. "Colour Schemes." SRON Technical Note, SRON/EPS/TN/09-002. https://sronpersonalpages.nl/~pault/data/colourschemes.pdf

See Also

Other colour ptol: scale_colour_ptol()

Examples

library("scales")

show_col(ptol_pal()(6))
show_col(ptol_pal()(4))
show_col(ptol_pal()(12))

Color scale from the pander package

Description

The pander ships with a default colorblind and printer-friendly color palette borrowed from ⁠https://jfly.iam.u-tokyo.ac.jp/color/⁠.

Usage

scale_color_pander(...)

scale_colour_pander(...)

scale_fill_pander(...)

Arguments

Value

A ggplot2 scale object.

See Also

theme_pander()

Other colour pander: palette_pander()

Examples

library("ggplot2")

ggplot(mtcars, aes(x = wt, y = mpg, colour = factor(gear))) +
  geom_point(size = 3) +
  scale_colour_pander()

Discrete color scale using canva.com color palettes

Description

Color scale for canva.com color palettes described in canva_palettes().

Usage

scale_colour_canva(..., palette = "Fresh and bright")

scale_color_canva(..., palette = "Fresh and bright")

scale_fill_canva(..., palette = "Fresh and bright")

Arguments

Value

A ggplot2 scale object.

Examples

library("ggplot2")

ggplot(mtcars, aes(x = wt, y = mpg, colour = factor(gear))) +
  geom_point(size = 3) +
  scale_colour_canva()

Economist color scales

Description

Color scales using the colors in the Economist graphics.

Usage

scale_colour_economist(...)

scale_color_economist(...)

scale_fill_economist(...)

Arguments

Value

A ggplot2 scale object.

See Also

theme_economist() for examples.

Other colour economist: economist_pal(), economist_seq_pal(), scale_colour_economist_c()

Examples

library("ggplot2")

ggplot(mtcars, aes(x = wt, y = mpg, colour = factor(gear))) +
  geom_point(size = 3) +
  scale_colour_economist()

Economist sequential color scales

Description

Color scales built from The Economist's equal-lightness color scales. The _c scales are continuous; the _ordinal scales are discrete, for ordered factors. See scale_colour_economist() for the unordered categorical scales.

Usage

scale_colour_economist_c(hue = "blue", guide = "colourbar", ...)

scale_color_economist_c(hue = "blue", guide = "colourbar", ...)

scale_fill_economist_c(hue = "blue", guide = "colourbar", ...)

scale_colour_economist_ordinal(hue = "blue", ...)

scale_color_economist_ordinal(hue = "blue", ...)

scale_fill_economist_ordinal(hue = "blue", ...)

Arguments

Value

A ggplot2 scale object.

See Also

Other colour economist: economist_pal(), economist_seq_pal(), scale_colour_economist()

Examples

library("ggplot2")

# Continuous scale
ggplot(mtcars, aes(x = wt, y = mpg, colour = disp)) +
  geom_point(size = 3) +
  scale_colour_economist_c(hue = "blue")

# Ordinal (discrete) scale for an ordered factor
ggplot(mtcars, aes(x = wt, y = mpg, colour = factor(cyl, ordered = TRUE))) +
  geom_point(size = 3) +
  scale_colour_economist_ordinal(hue = "red")

Excel (current versions) color scales

Description

Discrete color scales used in current versions of Microsoft Office and Excel.

Usage

scale_colour_excel_new(theme = "Office", ...)

scale_color_excel_new(theme = "Office", ...)

scale_fill_excel_new(theme = "Office", ...)

Arguments

Value

A ggplot2 scale object.

See Also

Other colour excel: excel_new_pal(), excel_pal(), scale_fill_excel()

Examples

library("ggplot2")

p <- ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, colour = factor(gear))) +
  facet_wrap(~am)
p + theme_excel_new() + scale_colour_excel_new()

Color scales from Few's "Practical Rules for Using Color in Charts"

Description

See few_pal().

Usage

scale_colour_few(palette = "Medium", ...)

scale_color_few(palette = "Medium", ...)

scale_fill_few(palette = "Light", ...)

Arguments

Value

A ggplot2 scale object.

See Also

Other colour few: few_pal()

Examples

library("ggplot2")

ggplot(mtcars, aes(x = wt, y = mpg, colour = factor(gear))) +
  geom_point(size = 3) +
  scale_colour_few("Dark")

FiveThirtyEight color scales

Description

Color scales using the colors in the FiveThirtyEight graphics.

Usage

scale_colour_fivethirtyeight(...)

scale_color_fivethirtyeight(...)

scale_fill_fivethirtyeight(...)

Arguments

Value

A ggplot2 scale object.

See Also

theme_fivethirtyeight() for examples.

Other colour fivethirtyeight: fivethirtyeight_pal()

Examples

library("ggplot2")

ggplot(mtcars, aes(x = wt, y = mpg, colour = factor(gear))) +
  geom_point(size = 3) +
  scale_colour_fivethirtyeight()

Tableau diverging colour scales (continuous)

Description

Continuous color scales using the diverging color scales in Tableau. See scale_colour_tableau() for Tabaleau discrete color scales, and scale_colour_gradient_tableau() for sequential color scales.

Usage

scale_colour_gradient2_tableau(
  palette = "Orange-Blue Diverging",
  ...,
  midpoint = 0,
  na.value = "grey50",
  guide = "colourbar"
)

scale_fill_gradient2_tableau(
  palette = "Orange-Blue Diverging",
  ...,
  midpoint = 0,
  na.value = "grey50",
  guide = "colourbar"
)

scale_color_gradient2_tableau(
  palette = "Orange-Blue Diverging",
  ...,
  midpoint = 0,
  na.value = "grey50",
  guide = "colourbar"
)

Arguments

Value

A ggplot2 scale object.

See Also

Other colour tableau: scale_colour_gradient_tableau(), scale_colour_tableau(), tableau_color_pal(), tableau_gradient_pal()

Examples

library("ggplot2")

df <- data.frame(
  x = runif(100),
  y = runif(100),
  z1 = rnorm(100),
  z2 = abs(rnorm(100))
)
p <- ggplot(df, aes(x, y)) +
  geom_point(aes(colour = z2))

palettes <-
  ggthemes_data[["tableau"]][["color-palettes"]][["ordered-diverging"]]
for (palette in head(names(palettes))) {
  print(p + scale_colour_gradient2_tableau(palette) + ggtitle(palette))
}

# If you need to reverse a palette, use a transformation
p + scale_colour_gradient2_tableau(trans = "reverse")

Tableau sequential colour scales (continuous)

Description

Continuous color scales using the sequential color palettes in Tableau. See scale_colour_tableau() for Tableau discrete color scales, and scale_colour_gradient2_tableau() for diverging color scales.

Usage

scale_colour_gradient_tableau(
  palette = "Blue",
  ...,
  na.value = "grey50",
  guide = "colourbar"
)

scale_fill_gradient_tableau(
  palette = "Blue",
  ...,
  na.value = "grey50",
  guide = "colourbar"
)

scale_color_gradient_tableau(
  palette = "Blue",
  ...,
  na.value = "grey50",
  guide = "colourbar"
)

scale_color_continuous_tableau(
  palette = "Blue",
  ...,
  na.value = "grey50",
  guide = "colourbar"
)

scale_fill_continuous_tableau(
  palette = "Blue",
  ...,
  na.value = "grey50",
  guide = "colourbar"
)

Arguments

Value

A ggplot2 scale object.

See Also

Other colour tableau: scale_colour_gradient2_tableau(), scale_colour_tableau(), tableau_color_pal(), tableau_gradient_pal()

Examples

library("ggplot2")

df <- data.frame(
  x = runif(100),
  y = runif(100),
  z1 = rnorm(100),
  z2 = abs(rnorm(100))
)

p <- ggplot(df, aes(x, y)) +
  geom_point(aes(colour = z2)) +
  theme_igray()

palettes <-
  ggthemes_data[["tableau"]][["color-palettes"]][["ordered-sequential"]]
for (palette in head(names(palettes))) {
  print(p + scale_colour_gradient_tableau(palette) + ggtitle(palette))
}

Highcharts color and fill scales

Description

Colour and fill scales which use the palettes in hc_pal() and are meant for use with theme_hc().

Usage

scale_colour_hc(palette = "default", ...)

scale_color_hc(palette = "default", ...)

scale_fill_hc(palette = "default", ...)

Arguments

Value

A ggplot2 scale object.

See Also

Other colour hc: hc_pal()

Examples

library("ggplot2")

ggplot(mtcars, aes(x = wt, y = mpg, colour = factor(gear))) +
  geom_point(size = 3) +
  scale_colour_hc(palette = "darkunica")

Color Scales from Paul Tol's "Colour Schemes

Description

These scales were deprecated in ggthemes 7.0.0. Use the khroma package instead, which tracks Paul Tol's colour schemes as he revises them.

They draw the 12-colour qualitative scheme from Tol's 2012 technical note, and have not followed the revisions he has made since. His current schemes are at https://sronpersonalpages.nl/~pault/; the closest successor is khroma::scale_colour_muted().

See ptol_pal(). These palettes support up to 12 values.

Usage

scale_colour_ptol(...)

scale_color_ptol(...)

scale_fill_ptol(...)

Arguments

Value

A ggplot2 scale object.

See Also

Other colour ptol: ptol_pal()

Examples

library("ggplot2")

p2 <- ggplot(mtcars, aes(x = wt, y = mpg, colour = factor(gear))) +
  geom_point() +
  geom_smooth(method = "lm", se = FALSE) +
  scale_color_ptol("cyl") +
  theme_minimal() +
  ggtitle("Cars")

ggplot(diamonds, aes(x = clarity, fill = cut)) +
  geom_bar() +
  scale_fill_ptol() +
  theme_minimal()

Stata color scales

Description

See stata_pal() for details.

Usage

scale_colour_stata(scheme = NULL, ...)

scale_fill_stata(scheme = NULL, ...)

scale_color_stata(scheme = NULL, ...)

Arguments

Value

A ggplot2 scale object.

See Also

Other colour stata: stata_pal()

Examples

library("ggplot2")

ggplot(mtcars, aes(x = wt, y = mpg, colour = factor(gear))) +
  geom_point(size = 3) +
  scale_colour_stata(scheme = "s2color")

Tableau color scales (discrete)

Description

Categorical (qualitative) color scales used in Tableau. Use the function scale_colour_gradient_tableau() for the sequential and scale_colour_gradient2_tableau() for the diverging continuous color scales from Tableau.

Usage

scale_colour_tableau(
  palette = "Tableau 10",
  type = "regular",
  direction = 1,
  ...
)

scale_fill_tableau(
  palette = "Tableau 10",
  type = "regular",
  direction = 1,
  ...
)

scale_color_tableau(
  palette = "Tableau 10",
  type = "regular",
  direction = 1,
  ...
)

Arguments

Value

A ggplot2 scale object.

See Also

tableau_color_pal() for references.

Other colour tableau: scale_colour_gradient2_tableau(), scale_colour_gradient_tableau(), tableau_color_pal(), tableau_gradient_pal()

Examples

library("ggplot2")

p <- ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, colour = factor(gear))) +
  facet_wrap(~am) +
  theme_igray()

palettes <- ggthemes_data[["tableau"]][["color-palettes"]][["regular"]]
for (palette in head(names(palettes), 3L)) {
  print(p + scale_colour_tableau(palette) + ggtitle(palette))
}

# the order of colour can be reversed
p + scale_color_tableau(direction = -1)

Wall Street Journal color and fill scales

Description

Colour and fill scales which use the palettes in wsj_pal(). These scales should be used with theme_wsj().

Usage

scale_colour_wsj(palette = "colors6", ...)

scale_color_wsj(palette = "colors6", ...)

scale_fill_wsj(palette = "colors6", ...)

Arguments

Value

A ggplot2 scale object.

See Also

Other colour wsj: wsj_pal()

Examples

library("ggplot2")

ggplot(mtcars, aes(x = wt, y = mpg, colour = factor(gear))) +
  geom_point(size = 3) +
  scale_colour_wsj("colors6")

LibreOffice Calc color scales

Description

Color scales from LibreOffice Calc.

Usage

scale_fill_calc(...)

scale_colour_calc(...)

scale_color_calc(...)

Arguments

Value

A ggplot2 scale object.

See Also

See theme_calc() for examples.

Other colour calc: calc_pal()

Examples

library("ggplot2")

ggplot(mtcars, aes(x = wt, y = mpg, colour = factor(gear))) +
  geom_point(size = 3) +
  scale_colour_calc()

Excel 97 ugly color scales

Description

The classic "ugly" color scales from Excel 97.

Usage

scale_fill_excel(...)

scale_colour_excel(...)

scale_color_excel(...)

Arguments

Value

A ggplot2 scale object.

See Also

Other colour excel: excel_new_pal(), excel_pal(), scale_colour_excel_new()

Examples

library("ggplot2")

# Line and scatter plot colors
p <- ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, colour = factor(gear))) +
  facet_wrap(~am)
p + theme_excel() + scale_colour_excel()

# Bar plot (area/fill) colors
ggplot(mpg, aes(x = class, fill = drv)) +
  geom_bar() +
  scale_fill_excel() +
  theme_excel()

Google Docs color scales

Description

Color scales from Google Docs.

Usage

scale_fill_gdocs(...)

scale_colour_gdocs(...)

scale_color_gdocs(...)

Arguments

Value

A ggplot2 scale object.

See Also

See theme_gdocs() for examples.

Other colour gdocs: gdocs_pal()

Examples

library("ggplot2")

ggplot(mtcars, aes(x = wt, y = mpg, colour = factor(gear))) +
  geom_point(size = 3) +
  scale_colour_gdocs()

Apple Numbers color scales

Description

Discrete color scales using the chart palettes from Apple Numbers.

Usage

scale_fill_numbers(palette = "Classic", ...)

scale_colour_numbers(palette = "Classic", ...)

scale_color_numbers(palette = "Classic", ...)

Arguments

Value

A ggplot2 scale object.

See Also

See theme_numbers() for examples.

Other colour numbers: numbers_pal()

Examples

library("ggplot2")

p <- ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, colour = factor(gear))) +
  facet_wrap(~am) +
  theme_numbers()

for (palette in head(names(ggthemes_data[["numbers"]]), 3L)) {
  print(p + scale_colour_numbers(palette) + ggtitle(palette))
}

Solarized color scales

Description

See solarized_pal() for details.

Usage

scale_fill_solarized(accent = "blue", ...)

scale_colour_solarized(accent = "blue", ...)

scale_color_solarized(accent = "blue", ...)

Arguments

Value

A ggplot2 scale object.

See Also

Other colour solarized: solarized_pal()

Examples

library("ggplot2")

p <- ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, colour = factor(gear))) +
  facet_wrap(~am)
p + theme_solarized() + scale_colour_solarized()

Stata linetype palette (discrete)

Description

See stata_linetype_pal() for details.

Usage

scale_linetype_stata(...)

Arguments

Value

A ggplot2 scale object.

See Also

Other linetype stata: stata_linetype_pal()

Examples

require("ggplot2")
if (require("tidyr") && require("dplyr")) {
  rescale01 <- function(x) {
    (x - min(x)) / diff(range(x))
  }

  gather(economics, variable, value, -date) |>
    group_by(variable) |>
    mutate(value = rescale01(value)) |>
    ggplot(aes(x = date, y = value, linetype = variable)) +
    geom_line() +
    scale_linetype_stata()
}

Calc shape scale

Description

See calc_shape_pal() for details.

Usage

scale_shape_calc(..., unicode = FALSE)

Arguments

Value

A ggplot2 scale object.

See Also

theme_calc() for examples.

Other shapes calc: calc_shape_pal()

Examples

library("ggplot2")

ggplot(mtcars, aes(x = wt, y = mpg, shape = factor(gear))) +
  geom_point(size = 3) +
  scale_shape_calc()

Filled Circle Shape palette (discrete)

Description

'r lifecycle::badge("deprecated")'

Usage

scale_shape_circlefill(...)

Arguments

Value

A ggplot2 scale object.

See Also

circlefill_shape_pal() for a description of the palette.

Other shapes: circlefill_shape_pal(), cleveland_shape_pal(), scale_shape_cleveland(), scale_shape_tremmel(), tremmel_shape_pal()

Examples

# Deprecated. The palette grades circles by how much of each is filled, which
# needs Unicode glyphs from the device font, so this example does not draw it.
scale_shape_circlefill()

Shape scales from Cleveland "Elements of Graphing Data"

Description

Shape scales from Cleveland "Elements of Graphing Data"

Usage

scale_shape_cleveland(overlap = TRUE, ..., unicode = FALSE)

Arguments

Value

A ggplot2 scale object.

References

Cleveland WS. The Elements of Graphing Data. Revised Edition. Hobart Press, Summit, NJ, 1994, pp. 154-164, 234-239.

See Also

cleveland_shape_pal() for a description of the palette.

Other shapes: circlefill_shape_pal(), cleveland_shape_pal(), scale_shape_circlefill(), scale_shape_tremmel(), tremmel_shape_pal()

Examples

library("ggplot2")

ggplot(mtcars, aes(x = wt, y = mpg, shape = factor(gear))) +
  geom_point(size = 3) +
  scale_shape_cleveland()

Scales for shapes from "Show Me the Numbers"

Description

scale_shape_few() maps discrete variables to up to five easily discernible shapes. It is based on the shape palette suggested in Few (2012).

Usage

scale_shape_few(...)

Arguments

Value

A ggplot2 scale object.

References

Few, S. (2012) Show Me the Numbers: Designing Tables and Graphs to Enlighten, Analytics Press, p. 208.

See Also

scale_shape_few() for the shape palette that this scale uses.

Examples

library("ggplot2")

ggplot(mtcars, aes(x = wt, y = mpg, shape = factor(gear))) +
  geom_point(size = 3) +
  scale_shape_few()

Stata shape scale

Description

See stata_shape_pal() for details.

Usage

scale_shape_stata(..., unicode = FALSE)

Arguments

Value

A ggplot2 scale object.

See Also

Other shapes stata: stata_shape_pal()

Examples

library("ggplot2")

p <- ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, shape = factor(gear))) +
  facet_wrap(~am)
p + theme_stata(scheme = "s2color") + scale_shape_stata()

Tableau shape scales

Description

See tableau_shape_pal() for details.

Usage

scale_shape_tableau(palette = "default", ..., unicode = FALSE)

Arguments

Value

A ggplot2 scale object.

See Also

Other shapes tableau: tableau_shape_pal()

Examples

library("ggplot2")

p <- ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, shape = factor(gear))) +
  facet_wrap(~am)
p + scale_shape_tableau()

Shape scales from Tremmel (1995)

Description

Shape scales from Tremmel (1995)

Usage

scale_shape_tremmel(overlap = FALSE, alt = FALSE, ...)

Arguments

Value

A ggplot2 scale object.

See Also

tremmel_shape_pal() for a description of the palette.

Other shapes: circlefill_shape_pal(), cleveland_shape_pal(), scale_shape_circlefill(), scale_shape_cleveland(), tremmel_shape_pal()

Examples

library("ggplot2")

p <- ggplot(mtcars, aes(x = mpg, y = hp, shape = factor(cyl))) +
  geom_point()

p + scale_shape_tremmel()
p + scale_shape_tremmel(alt = TRUE)
p + scale_shape_tremmel(overlap = TRUE)

Show linetypes

Description

A quick and dirty way to show linetypes.

Usage

show_linetypes(linetypes, labels = TRUE)

Arguments

Value

This function called for the side effect of creating a plot. It returns linetypes.

See Also

show_col(), show_linetypes()

Examples

library("scales")

show_linetypes(linetype_pal()(3))
show_linetypes(linetype_pal()(3), labels = TRUE)

Show shapes

Description

A quick and dirty way to show shapes.

Usage

show_shapes(shapes, labels = TRUE)

Arguments

Value

This function called for the side effect of creating a plot. It returns shapes.

See Also

show_col(), show_linetypes()

Examples

library("scales")

show_shapes(shape_pal()(5))
show_shapes(shape_pal()(3), labels = TRUE)

Format numbers with automatic number of digits

Description

Format numbers with automatic number of digits

Usage

smart_digits(x, ...)

smart_digits_format(x, ...)

Arguments

Value

A character vector. smart_digits_format() returns a function with a single argument x, a numeric vector, that returns a character vector.

Author(s)

Josh O'Brien, Baptiste Auguie, Jeffrey B. Arnold

References

Josh O'Brien, https://stackoverflow.com/questions/23169938/select-accuracy-to-display-additional-axis-breaks/23171858#23171858.

Examples

smart_digits(c(0.1234, 0.5, 1.25))

smart_digits(c(1234.5678, 2000, 10000))

# A labelling function for use in a scale
library("ggplot2")

ggplot(mtcars, aes(x = wt, y = mpg)) +
  geom_point() +
  scale_y_continuous(labels = smart_digits_format())

Solarized color palette (discrete)

Description

Qualitative color palate based on the Ethan Schoonover's Solarized palette, https://ethanschoonover.com/solarized/. This palette supports up to seven values.

Usage

solarized_pal(accent = "blue")

Arguments

Value

A palette function. It takes the number of colours n and returns a character vector of n hex colours, and can be used as the palette argument of discrete_scale().

Note

For a given starting color and number of colors in the palette, the other colors are the combination of colors that maximizes the total Euclidean distance between colors in L*a*b space.

See Also

Other colour solarized: scale_fill_solarized()

Examples

library("scales")

show_col(solarized_pal()(2))
show_col(solarized_pal()(3))
show_col(solarized_pal("red")(4))

Base colors for Solarized light and dark themes

Description

Base colors for Solarized light and dark themes

Usage

solarized_rebase(light = TRUE)

Arguments

Value

A named character vector of eight hex colours, named rebase03 to rebase3.

Calculate components of a five-number summary

Description

The five number summary of a sample is the minimum, first quartile, median, third quartile, and maximum.

Usage

stat_fivenumber(
  mapping = NULL,
  data = NULL,
  geom = "boxplot",
  probs = c(0, 0.25, 0.5, 0.75, 1),
  na.rm = FALSE,
  position = "identity",
  show.legend = NA,
  inherit.aes = TRUE,
  ...
)

Arguments

Value

A data frame with additional columns:

See Also

stat_boxplot()

Examples

library("ggplot2")

ggplot(mtcars, aes(x = factor(cyl), y = mpg)) +
  stat_fivenumber()

# The whiskers can be set to other quantiles
ggplot(mtcars, aes(x = factor(cyl), y = mpg)) +
  stat_fivenumber(probs = c(0.05, 0.25, 0.5, 0.75, 0.95))

Stata linetype palette (discrete)

Description

Linetype palette based on the linepattern scheme in Stata. This palette supports up to 15 values.

Usage

stata_linetype_pal()

Value

A palette function. It takes the number of linetypes n and returns a character vector of n linetype specifications, and can be used as the palette argument of discrete_scale().

See Also

scale_linetype_stata()

Other linetype stata: scale_linetype_stata()

Examples

stata_linetype_pal()(6)

show_linetypes(stata_linetype_pal()(6))

Stata color palettes (discrete)

Description

Stata color palettes. See Stata documentation for a description of the schemes, https://www.stata.com/help.cgi?schemes.

Usage

stata_pal(scheme = NULL)

Arguments

Details

All these palettes support up to 15 values.

Stata's palettes come in two generations, and both are included here.

Stata 17 and earlier

Schemes "s2color", "s1color", "s1rcolor", and "mono" are built from Stata's classic named colors (navy, maroon, forest_green, and so on) and the gs0–gs16 gray scale. "s2color" was Stata's factory default through Stata 17.

Stata 18 and later

Scheme "stcolor" uses the stc1–stc15 colors introduced in Stata 18. They are brighter than the classic palette and chosen to stay distinguishable for readers with a color vision deficiency. The first four are also available under the aliases stblue, stred, stgreen, and styellow. "stcolor" has been Stata's factory default since Stata 18.

"economist" is not one of Stata's general-purpose schemes; it is the set of Economist-styled colors that Stata ships in scheme-economist.scheme.

Value

A palette function. It takes the number of colours n and returns a character vector of n hex colours, and can be used as the palette argument of discrete_scale().

See Also

Other colour stata: scale_colour_stata()

Examples

library("scales")

# Stata 18 and later (the current factory default)
show_col(stata_pal("stcolor")(15))

# Stata 17 and earlier
show_col(stata_pal("s2color")(15))
show_col(stata_pal("s1rcolor")(15))
show_col(stata_pal("s1color")(15))
show_col(stata_pal("mono")(15))

Stata shape palette (discrete)

Description

Shape palette based on the symbol palette in Stata used in scheme s2mono. This palette supports up to 10 values.

Usage

stata_shape_pal(unicode = FALSE)

Arguments

Value

A palette function. It takes the number of shapes n and returns an integer vector of n shape (pch) codes, and can be used as the palette argument of discrete_scale().

Note

Stata's ten plotting symbols all have a base pch equivalent, so this palette supports ten values on either branch and nothing is dropped: solid and hollow circle, diamond, square and triangle, plus the X and the plus sign.

See Also

See scale_shape_stata() for examples.

Other shapes stata: scale_shape_stata()

Examples

stata_shape_pal()(10)

show_shapes(stata_shape_pal()(10))

Tableau Color Palettes (discrete)

Description

Color palettes used in Tableau.

Usage

tableau_color_pal(
  palette = "Tableau 10",
  type = c("regular", "ordered-sequential", "ordered-diverging"),
  direction = 1
)

Arguments

Details

Tableau provides three types of color palettes: "regular" (discrete, qualitative categories), "ordered-sequential", and "ordered-diverging".

"regular"

"Tableau 10", "Tableau 20", "Color Blind", "Seattle Grays", "Traffic", "Miller Stone", "Superfishel Stone", "Nuriel Stone", "Jewel Bright", "Summer", "Winter", "Green-Orange-Teal", "Blue-Red-Brown", "Purple-Pink-Gray", "Hue Circle", "Classic 10", "Classic 10 Medium", "Classic 10 Light", "Classic 20", "Classic Gray 5", "Classic Color Blind", "Classic Traffic Light", "Classic Purple-Gray 6", "Classic Purple-Gray 12", "Classic Green-Orange 6", "Classic Green-Orange 12", "Classic Blue-Red 6", "Classic Blue-Red 12", "Classic Cyclic"

"ordered-diverging"

"Orange-Blue Diverging", "Red-Green Diverging", "Green-Blue Diverging", "Red-Blue Diverging", "Red-Black Diverging", "Gold-Purple Diverging", "Red-Green-Gold Diverging", "Sunset-Sunrise Diverging", "Orange-Blue-White Diverging", "Red-Green-White Diverging", "Green-Blue-White Diverging", "Red-Blue-White Diverging", "Red-Black-White Diverging", "Orange-Blue Light Diverging", "Temperature Diverging", "Classic Red-Green", "Classic Red-Blue", "Classic Red-Black", "Classic Area Red-Green", "Classic Orange-Blue", "Classic Green-Blue", "Classic Red-White-Green", "Classic Red-White-Black", "Classic Orange-White-Blue", "Classic Red-White-Black Light", "Classic Orange-White-Blue Light", "Classic Red-White-Green Light", "Classic Red-Green Light"

"ordered-sequential"

"Blue-Green Sequential", "Blue Light", "Orange Light", "Blue", "Orange", "Green", "Red", "Purple", "Brown", "Gray", "Gray Warm", "Blue-Teal", "Orange-Gold", "Green-Gold", "Red-Gold", "Classic Green", "Classic Gray", "Classic Blue", "Classic Red", "Classic Orange", "Classic Area Red", "Classic Area Green", "Classic Area Brown"

Value

A palette function. It takes the number of colours n and returns a character vector of n hex colours, and can be used as the palette argument of discrete_scale().

References

http://vis.stanford.edu/color-names/analyzer/

Maureen Stone, 'Designing Colors for Data' (slides), at the International Symposium on Computational Aesthetics in Graphics, Visualization, and Imaging, Banff, AB, Canada, June 22, 2007.

Heer, Jeffrey and Maureen Stone, 2012 'Color Naming Models for Color Selection, Image Editing and Palette Design', ACM Human Factors in Computing Systems (CHI) http://vis.stanford.edu/files/2012-ColorNameModels-CHI.pdf.

See Also

Other colour tableau: scale_colour_gradient2_tableau(), scale_colour_gradient_tableau(), scale_colour_tableau(), tableau_gradient_pal()

Examples

library("scales")

palettes <- ggthemes_data[["tableau"]][["color-palettes"]][["regular"]]
for (palname in names(palettes)) {
  pal <- tableau_color_pal(palname)
  max_n <- attr(pal, "max_n")
  show_col(pal(max_n))
  title(main = palname)
}

Tableau colour gradient palettes (continuous)

Description

Gradient color palettes using the diverging and sequential continous color palettes in Tableau. See tableau_color_pal() for discrete color palettes.

Usage

tableau_gradient_pal(palette = "Blue", type = "ordered-sequential")

tableau_seq_gradient_pal(palette = "Blue", ...)

tableau_div_gradient_pal(palette = "Orange-Blue Diverging", ...)

Arguments

Value

A palette function. It takes a numeric vector x of values between 0 and 1 and returns a character vector of hex colours interpolated along the palette, for use as the palette argument of continuous_scale().

See Also

Other colour tableau: scale_colour_gradient2_tableau(), scale_colour_gradient_tableau(), scale_colour_tableau(), tableau_color_pal()

Examples

library("scales")

x <- seq(0, 1, length = 25)
r <- sqrt(outer(x^2, x^2, "+"))
palettes <-
  ggthemes_data[["tableau"]][["color-palettes"]][["ordered-sequential"]]
for (palname in names(palettes)) {
  col <- tableau_seq_gradient_pal(palname)(seq(0, 1, length = 12))
  image(r, col = col)
  title(main = palname)
}

Tableau Shape Palettes (discrete)

Description

Shape palettes used by Tableau.

Usage

tableau_shape_pal(
  palette = c("default", "filled", "proportions"),
  unicode = FALSE
)

Arguments

Details

Not all shape palettes in Tableau are supported, and these palettes are not exact.

Shape palettes in Tableau are used to expose images for use as markers in charts, and thus are sometimes groupings of closely related symbols.

Value

A palette function. It takes the number of shapes n and returns an integer vector of n shape (pch) codes, and can be used as the palette argument of discrete_scale().

Note

Supported values by palette: "default" eight (ten with unicode = TRUE), "filled" six (ten), "proportions" two (five). Shapes with no base pch equivalent – the sideways triangles, the solid star, and the partially filled circles – are dropped rather than approximated by a different shape.

"proportions" encodes fill fraction, which base pch cannot express at all, so only its empty and full circles survive; they remain meaningful as a two-value scale. To encode a proportion, map alpha or fill instead, or use unicode = TRUE with a font covering Geometric Shapes, such as DejaVu Sans.

See Also

Other shapes tableau: scale_shape_tableau()

Examples

# The eight shapes with a font-independent equivalent.
show_shapes(tableau_shape_pal()(8))

## Not run: 
  # All ten, drawn from the device font. Needs a font covering Geometric
  # Shapes, such as DejaVu Sans.
  show_shapes(tableau_shape_pal(unicode = TRUE)(10))

## End(Not run)

Theme Base

Description

Theme similar to the default settings of the ‘base’ R graphics.

Usage

theme_base(base_size = 16, base_family = "")

Arguments

Value

A ggplot2 theme object (class theme).

See Also

Other themes: theme_clean(), theme_foundation(), theme_igray(), theme_par(), theme_solid()

Examples

library("ggplot2")

p <- ggplot(mtcars) +
  geom_point(aes(
    x = wt,
    y = mpg,
    colour = factor(gear)
  )) +
  facet_wrap(~am)
p + theme_base()
# Change values of par
old_par <- par(fg = "blue", bg = "gray", col.lab = "red", font.lab = 3)
p + theme_base()
par(old_par)

Theme Calc

Description

Theme similar to the default settings of LibreOffice Calc charts.

Usage

theme_calc(base_size = 10, base_family = "sans")

Arguments

Value

A ggplot2 theme object (class theme).

Examples

library("ggplot2")

ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, colour = factor(gear))) +
  facet_wrap(~am) +
  theme_calc() +
  scale_color_calc()

ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, shape = factor(gear))) +
  facet_wrap(~am) +
  theme_calc() +
  scale_shape_calc()

Clean ggplot theme

Description

Clean ggplot theme with no panel background, black axis lines and grey fill colour for chart elements.

Usage

theme_clean(base_size = 12, base_family = "sans")

Arguments

Value

A ggplot2 theme object (class theme).

Author(s)

Konrad Zdeb name.surname@me.com

See Also

Other themes: theme_base(), theme_foundation(), theme_igray(), theme_par(), theme_solid()

Examples

library("ggplot2")
p <- ggplot(mtcars, aes(x = wt, y = mpg, colour = factor(gear))) +
  geom_point() +
  facet_wrap(~am) +
  geom_smooth(method = "lm", se = FALSE) +
  theme_clean()
p

ggplot color theme based on the Economist

Description

A theme that approximates the style of charts in The Economist.

Usage

theme_economist(
  base_size = 10,
  base_family = "sans",
  horizontal = TRUE,
  dkpanel = deprecated()
)

theme_economist_white(
  base_size = 10,
  base_family = "sans",
  gray_bg = deprecated(),
  horizontal = TRUE
)

Arguments

Details

This follows the chart design The Economist introduced in 2017 and still publishes today: a white plot area on a pale ground, light horizontal gridlines only, a black x-axis baseline with tick marks below it, and no y-axis rule or ticks. Use scale_colour_economist() with it.

Two conventions of the house style cannot be expressed as theme elements, and have to be set on the plot itself:

• The Economist puts the y axis on the right. Use scale_y_continuous(position = "right").

• Charts are tagged with a small red rectangle above the title. Draw it with annotate() or grid.rect() in "Econ red", which is ggthemes_data$economist$main row "econ red".

The Economist sets charts in "Econ Sans", which is not publicly available. Any narrow humanist sans is a reasonable substitute; with the extrafont package, "Roboto Condensed" or "Fira Sans Condensed" are close.

Value

An object of class theme().

References

• The Economist

• The Economist visual styleguide, version 1.2, 4 May 2017.

Examples

library("ggplot2")

p <- ggplot(mtcars) +
     geom_point(aes(x = wt, y = mpg, colour = factor(gear))) +
     facet_wrap(~am) +
     # The Economist puts the y-axis labels on the right-hand side
     scale_y_continuous(position = "right") +
     labs(
       title = "Heavier, thirstier",
       subtitle = "Fuel economy v weight, by number of forward gears",
       caption = "Source: Motor Trend, 1974"
     )

## Standard
p + theme_economist() +
  scale_colour_economist()

# Vertical gridlines, for use with coord_flip()
p + theme_economist(horizontal = FALSE) +
    scale_colour_economist() +
    coord_flip()

## Ordered data uses one hue's equal-lightness steps instead
ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, colour = hp)) +
  scale_colour_economist_c(hue = "blue") +
  theme_economist()

## Not run: 

## The Economist sets charts in "Econ Sans", which is not publicly
## available. Any narrow humanist sans is a reasonable substitute, if it is
## installed on your system.
p + theme_economist(base_family = "Roboto Condensed") +
    scale_colour_economist()


## End(Not run)

ggplot theme based on old Excel plots

Description

Theme to replicate the ugly monstrosity that was the old gray-background Excel chart. Please never use this. This theme should be combined with the scale_colour_excel() color scale.

Usage

theme_excel(base_size = 12, base_family = "", horizontal = TRUE)

Arguments

Value

An object of class theme().

See Also

Other themes excel: theme_excel_new()

Examples

library("ggplot2")

# Line and scatter plot colors
p <- ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, colour = factor(gear))) +
  facet_wrap(~am)
p + theme_excel() + scale_colour_excel()

# Bar plot (area/fill) colors
ggplot(mpg, aes(x = class, fill = drv)) +
  geom_bar() +
  scale_fill_excel() +
  theme_excel()

ggplot theme similar to current Excel plot defaults

Description

Theme for ggplot2 that is similar to the default style of charts in current versions of Microsoft Excel.

Usage

theme_excel_new(base_size = 9, base_family = "sans")

Arguments

Details

Excel derives its chart greys from the theme's tx1 colour by luminance transform rather than hardcoding them. Since tx1 is black in every built-in Office theme, these greys—"#D9D9D9" gridlines, "#BFBFBF" axis lines, "#595959" text—are the same whichever theme scale_colour_excel_new() is set to.

Since 2023 the default font in Excel has been Aptos, but base_family defaults to "sans" because Aptos is rarely installed outside of Office. Pass base_family = "Aptos Narrow" for a closer match if you do have it.

Value

An object of class theme().

See Also

Other themes excel: theme_excel()

Examples

library("ggplot2")

p <- ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, colour = factor(gear))) +
  facet_wrap(~am)
p + theme_excel_new() + scale_colour_excel_new()

Theme based on Few's "Practical Rules for Using Color in Charts"

Description

Theme based on the rules and examples from Stephen Few's Show Me the Numbers and "Practical Rules for Using Color in Charts".

Usage

theme_few(base_size = 12, base_family = "")

Arguments

Value

A ggplot2 theme object (class theme).

References

Few, S. (2012) Show Me the Numbers: Designing Tables and Graphs to Enlighten. 2nd edition. Analytics Press.

Stephen Few, "Practical Rules for Using Color in Charts", https://www.perceptualedge.com/articles/visual_business_intelligence/rules_for_using_color.pdf.

Examples

library("ggplot2")

p <- ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, colour = factor(gear))) +
  facet_wrap(~am)
p + theme_few() + scale_colour_few()
p + theme_few() + scale_colour_few("Light")
p + theme_few() + scale_colour_few("Dark")

ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, shape = factor(gear))) +
  theme_few() +
  scale_shape_few()

Theme inspired by FiveThirtyEight plots

Description

Theme inspired by the plots from FiveThirtyEight.com.

Usage

theme_fivethirtyeight(base_size = 12, base_family = "sans")

Arguments

Value

A ggplot2 theme object (class theme).

Examples

library("ggplot2")
p <- ggplot(mtcars, aes(x = wt, y = mpg, colour = factor(gear))) +
  geom_point() +
  facet_wrap(~am) +
  geom_smooth(method = "lm", se = FALSE) +
  scale_color_fivethirtyeight() +
  theme_fivethirtyeight()
p

Foundation Theme

Description

This theme is designed to be a foundation from which to build new themes, and not meant to be used directly. theme_foundation() is a complete theme with only minimal number of elements defined. It is easier to create new themes by extending this one rather than theme_gray() or theme_bw(), because those themes define elements deep in the hierarchy.

Usage

theme_foundation(
  base_size = 12,
  base_family = "",
  ink = "black",
  paper = "white",
  accent = "#3366FF"
)

Arguments

Details

This theme takes theme_gray() and sets all colour and fill values to NULL, except for the top-level elements (line, rect, and title), which have colour = ink, and fill = paper. This leaves the spacing and-non colour defaults of the default ggplot2 themes in place.

Unlike theme_foundation(), the other themes in this package (e.g. theme_economist(), theme_excel(), theme_hc()) intentionally replicate a fixed, published visual style, so they do not expose ink/paper/accent arguments.

Value

A ggplot2 theme object (class theme).

See Also

Other themes: theme_base(), theme_clean(), theme_igray(), theme_par(), theme_solid()

Examples

library("ggplot2")

ggplot(mtcars, aes(x = wt, y = mpg)) +
  geom_point() +
  theme_foundation()

# Extend the foundation to build a new theme
theme_minimal_box <- function(...) {
  theme_foundation(...) +
    theme(
      panel.grid = element_blank(),
      axis.ticks = element_line()
    )
}

ggplot(mtcars, aes(x = wt, y = mpg)) +
  geom_point() +
  theme_minimal_box()

Theme with Google Docs Chart defaults

Description

Theme similar to the default look of charts in Google Docs.

Usage

theme_gdocs(base_size = 12, base_family = "sans")

Arguments

Value

A ggplot2 theme object (class theme).

Examples

library("ggplot2")

p <- ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, colour = factor(gear))) +
  facet_wrap(~am)
p + theme_gdocs() + scale_color_gdocs()

Highcharts Theme

Description

Themes based on Highcharts plots.

Usage

theme_hc(
  base_size = 12,
  base_family = "sans",
  style = c("default", "default_dark", "darkunica", "grid_light", "sand_signika"),
  bgcolor = NULL
)

Arguments

Details

Only the Highcharts themes that restyle the chart itself get a style here. The "high-contrast", "avocado" and "sunset" themes shipped with Highcharts 13 change nothing but the series colours, so they are available through hc_pal() alone; combine them with theme_hc("default") or theme_hc("default_dark").

Highcharts pairs several of these themes with a web font (darkunica with Unica One, grid_light with Dosis, sand_signika with Signika). Those are not requested here, since the font may not be installed; pass base_family to use one.

Value

A ggplot2 theme object (class theme).

References

https://www.highcharts.com/demo/highcharts/line-chart

Examples

library("ggplot2")

p <- ggplot(mtcars) +
  geom_point(aes(
    x = wt,
    y = mpg,
    colour = factor(gear)
  )) +
  facet_wrap(~am)
p + theme_hc() + scale_colour_hc()
p + theme_hc(style = "darkunica") + scale_colour_hc("darkunica")
p + theme_hc(style = "grid_light") + scale_colour_hc("grid_light")
p + theme_hc(style = "default_dark") + scale_colour_hc("high_contrast_dark")

dtemp <- data.frame(
  months = factor(rep(substr(month.name, 1, 3), 4), levels = substr(month.name, 1, 3)),
  city = rep(c("Tokyo", "New York", "Berlin", "London"), each = 12),
  temp = c(
    7.0,
    6.9,
    9.5,
    14.5,
    18.2,
    21.5,
    25.2,
    26.5,
    23.3,
    18.3,
    13.9,
    9.6,
    -0.2,
    0.8,
    5.7,
    11.3,
    17.0,
    22.0,
    24.8,
    24.1,
    20.1,
    14.1,
    8.6,
    2.5,
    -0.9,
    0.6,
    3.5,
    8.4,
    13.5,
    17.0,
    18.6,
    17.9,
    14.3,
    9.0,
    3.9,
    1.0,
    3.9,
    4.2,
    5.7,
    8.5,
    11.9,
    15.2,
    17.0,
    16.6,
    14.2,
    10.3,
    6.6,
    4.8
  )
)

ggplot(dtemp, aes(x = months, y = temp, group = city, color = city)) +
  geom_line() +
  geom_point(size = 1.1) +
  ggtitle("Monthly Average Temperature") +
  theme_hc() +
  scale_colour_hc()

ggplot(dtemp, aes(x = months, y = temp, group = city, color = city)) +
  geom_line() +
  geom_point(size = 1.1) +
  ggtitle("Monthly Average Temperature") +
  theme_hc(style = "darkunica") +
  scale_colour_hc("darkunica")

Inverse gray theme

Description

Theme with white panel and gray background.

Usage

theme_igray(base_size = 12, base_family = "")

Arguments

Value

A ggplot2 theme object (class theme).

Details

This theme inverts the colors in the theme_gray(), a white panel and a light gray area around it. This keeps a white background for the color scales like theme_bw(). But by using a gray background, the plot is closer to the typographical color of the document, which is the motivation for using a gray panel in theme_gray(). This is similar to the style of plots in Stata and Tableau.

See Also

theme_gray(), theme_bw()

Other themes: theme_base(), theme_clean(), theme_foundation(), theme_par(), theme_solid()

Examples

library("ggplot2")

p <- ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, colour = factor(gear))) +
  facet_wrap(~am)
p + theme_igray()

Clean theme for maps

Description

A clean theme that is good for displaying maps from geom_map().

Usage

theme_map(base_size = 9, base_family = "")

Arguments

Value

A ggplot2 theme object (class theme).

Examples

library("ggplot2")

if (requireNamespace("maps", quietly = TRUE) && requireNamespace("mapproj", quietly = TRUE)) {
  us <- map_data("state")
  gg <- ggplot(us, aes(x = long, y = lat, group = group)) +
    geom_polygon(fill = "white", color = "black", linewidth = 0.25) +
    coord_map("albers", lat0 = 39, lat1 = 45) +
    theme_map()
  gg
}

Theme with Apple Numbers chart defaults

Description

Theme similar to the default look of charts in Apple Numbers.

Usage

theme_numbers(base_size = 12, base_family = "sans")

Arguments

Details

The values used here are those in the chart-style-default style of the theme stylesheet that ships inside Numbers: no chart background fill, gridlines in the value direction only, a border along the bottom of the chart but not the other three sides, and no tick marks.

Value

A ggplot2 theme object (class theme).

Examples

library("ggplot2")

p <- ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, colour = factor(gear))) +
  facet_wrap(~am)
p + theme_numbers() + scale_color_numbers()

A ggplot theme originated from the pander package

Description

The pander ships with a default theme when the 'unify plots' option is enabled via panderOptions, which is now also available outside of pander internals, like evals, eval.msgs or Pandoc.brew.

Usage

theme_pander(
  base_size = 12,
  base_family = "sans",
  nomargin = TRUE,
  ff = NULL,
  fc = "black",
  fs = NULL,
  gM = TRUE,
  gm = TRUE,
  gc = "grey",
  gl = "dashed",
  boxes = FALSE,
  bc = "white",
  pc = "transparent",
  lp = "right",
  axis = 1
)

Arguments

Value

A ggplot2 theme object (class theme).

Examples

require("ggplot2")
if (require("pander")) {
  p <- ggplot(mtcars, aes(x = mpg, y = wt)) +
    geom_point()
  p + theme_pander()

  old_grid_color <- panderOptions("graph.grid.color")
  panderOptions("graph.grid.color", "red")
  p + theme_pander()
  panderOptions("graph.grid.color", old_grid_color)

  p <- ggplot(mtcars, aes(wt, mpg, colour = factor(cyl))) +
    geom_point()
  p + theme_pander() + scale_color_pander()

  ggplot(mpg, aes(x = class, fill = drv)) +
    geom_bar() +
    scale_fill_pander() +
    theme_pander()
}

Theme which uses the current ‘base’ graphics parameter values from par(). Not all par() parameters, are supported, and not all are relevant to ggplot2 themes.

Description

Currently this theme uses the values of the parameters: "code", ""ps"", "code" "family", "fg", "bg", "adj", "font", "cex.axis", "cex.lab", "cex.main", "cex.sub", "col.axis", "col.lab", "col.main", "col.sub", "font", "font.axis", "font.lab", "font.main", "font.sub", "las", "lend", "lheight", "lty", "mar", "ps", "tcl", "tck", "xaxt", "yaxt".

Usage

theme_par(base_size = par()$ps, base_family = par()$family)

Arguments

Details

This theme does not translate the base graphics perfectly, so the graphs produced by it will not be identical to those produced by base graphics, most notably in the spacing of the margins.

Value

A ggplot2 theme object (class theme).

See Also

Other themes: theme_base(), theme_clean(), theme_foundation(), theme_igray(), theme_solid()

Examples

library("ggplot2")

p <- ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, colour = factor(gear))) +
  facet_wrap(~am)

p + theme_par()

# theme changes with respect to values of par
old_par <- par(font = 2, col.lab = "red", fg = "white", bg = "black")
p + theme_par()
par(old_par)

ggplot color themes based on the Solarized palette

Description

See https://ethanschoonover.com/solarized/ for a description of the Solarized palette.

Usage

theme_solarized(base_size = 12, base_family = "", light = TRUE)

theme_solarized_2(base_size = 12, base_family = "", light = TRUE)

Arguments

Details

Plots made with this theme integrate seamlessly with the Solarized Beamer color theme. https://github.com/jrnold/beamercolorthemesolarized. There are two variations: theme_solarized is similar to to theme_bw(), while theme_solarized_2() is similar to theme_gray().

Value

A ggplot2 theme object (class theme).

Examples

library("ggplot2")

p <- ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, colour = factor(gear)))

# Light version with different main accent colors
for (accent in names(ggthemes::ggthemes_data[["solarized"]][["accents"]])) {
  print(p + theme_solarized() + scale_colour_solarized(accent))
}

# Dark version
p + theme_solarized(light = FALSE) + scale_colour_solarized("blue")

# Alternative theme
p + theme_solarized_2(light = FALSE) + scale_colour_solarized("blue")

Theme with nothing other than a background color

Description

Theme that removes all non-geom elements (lines, text, etc), This theme is when only the geometric objects are desired.

Usage

theme_solid(base_size = 12, base_family = "", fill = NA)

Arguments

Value

A ggplot2 theme object (class theme).

See Also

Other themes: theme_base(), theme_clean(), theme_foundation(), theme_igray(), theme_par()

Examples

library("ggplot2")

ggplot(mtcars, aes(wt, mpg)) +
  geom_point() +
  theme_solid(fill = "white")

ggplot(mtcars, aes(wt, mpg)) +
  geom_point(color = "white") +
  theme_solid(fill = "black")

Themes based on Stata graph schemes

Description

Themes based on Stata graph schemes

Usage

theme_stata(base_size = 11, base_family = "sans", scheme = NULL)

Arguments

Details

These themes approximate Stata schemes using the features ggplot2. The graphical models of Stata and ggplot2 differ in various ways that make an exact replication impossible (or more difficult than it is worth). Some features in Stata schemes not in ggplot2: defaults for specific graph types, different levels of titles, captions and notes. These themes also adopt some of the ggplot2 defaults, and more effort was made to match the colors and sizes of major elements than in matching the margins.

The schemes fall into two generations. "stcolor", "stcolor_alt", "stmono1", "stmono2" and "stsj" are the st family introduced in Stata 18, of which "stcolor" is Stata's current factory default: a white background, a dashed grid on both axes, horizontal y-axis labels, and a borderless legend beside the plot. The remaining schemes are the s1/s2 families that were the default through Stata 17.

Stata expresses text sizes as a percentage of graph height, while ggplot2 uses absolute points, so the two agree only at a particular graph size. The relative sizes here match Stata exactly; base_size = 12.4 reproduces Stata's absolute sizes at its default 7.5 by 4.5 inch graph. Two further differences are not expressible in a ggplot2 theme: the number of legend columns (set by guide_legend() rather than the theme) and Stata's small default marker size (a geom default).

Value

A ggplot2 theme object (class theme).

References

https://www.stata.com/help.cgi?schemes

Examples

library("ggplot2")

p <- ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, colour = factor(gear))) +
  facet_wrap(~am) +
  labs(
    title = "Graphs by car type",
    x = "Weight (lbs.)",
    y = "MPG"
  )

# The st family, Stata's default since Stata 18
# stcolor
p + theme_stata(scheme = "stcolor") + scale_colour_stata("stcolor")
# stcolor_alt, which puts the legend below the plot
p + theme_stata(scheme = "stcolor_alt") + scale_colour_stata("stcolor")
# stmono1
p + theme_stata(scheme = "stmono1") + scale_colour_stata("mono")
# stmono2
p + theme_stata(scheme = "stmono2") + scale_colour_stata("mono")
# stsj, the Stata Journal scheme
p + theme_stata(scheme = "stsj") + scale_colour_stata("mono")

# The s1/s2 families, Stata's defaults through Stata 17
# s2color
p + theme_stata(scheme = "s2color") + scale_colour_stata("s2color")
# s2mono
p + theme_stata(scheme = "s2mono") + scale_colour_stata("mono")
# s1color
p + theme_stata(scheme = "s1color") + scale_colour_stata("s1color")
# s1rcolor
p + theme_stata(scheme = "s1rcolor") + scale_colour_stata("s1rcolor")
# s1mono
p + theme_stata(scheme = "s1mono") + scale_colour_stata("mono")

Tufte Maximal Data, Minimal Ink Theme

Description

Theme based on Chapter 6 'Data-Ink Maximization and Graphical Design' of Edward Tufte *The Visual Display of Quantitative Information*. No border, no axis lines, no grids. This theme works best in combination with geom_rug() or geom_rangeframe().

Usage

theme_tufte(base_size = 11, base_family = "serif", ticks = TRUE)

Arguments

Value

A ggplot2 theme object (class theme).

Note

The default font family is set to 'serif' as he uses serif fonts for labels in 'The Visual Display of Quantitative Information'. The serif font used by Tufte in his books is a variant of Bembo, while the sans serif font is Gill Sans. If these fonts are installed on your system, then you can use them with the package extrafont.

References

Tufte, Edward R. (2001) The Visual Display of Quantitative Information, Chapter 6.

See Also

Other tufte: extended_range_breaks_(), geom_rangeframe(), geom_tufteboxplot()

Examples

library("ggplot2")

p <- ggplot(mtcars, aes(x = wt, y = mpg)) +
  geom_point() +
  scale_x_continuous(breaks = extended_range_breaks()(mtcars$wt)) +
  scale_y_continuous(breaks = extended_range_breaks()(mtcars$mpg)) +
  ggtitle("Cars")

p + geom_rangeframe() + coord_cartesian(clip = "off") + theme_tufte()

p + geom_rug() + theme_tufte(ticks = FALSE)

Wall Street Journal theme

Description

Theme based on the plots in The Wall Street Journal.

Usage

theme_wsj(
  base_size = 12,
  color = "brown",
  base_family = "sans",
  title_family = "mono"
)

Arguments

Details

This theme should be used with scale_color_wsj().

Value

A ggplot2 theme object (class theme).

References

https://x.com/WSJGraphics

https://pinterest.com/wsjgraphics/wsj-graphics/

Examples

library("ggplot2")

p <- ggplot(mtcars) +
  geom_point(aes(x = wt, y = mpg, colour = factor(gear))) +
  facet_wrap(~am) +
  ggtitle("Diamond Prices")
p + scale_colour_wsj("colors6", "") + theme_wsj()
# Use a gray background instead
p + scale_colour_wsj("colors6", "") + theme_wsj(color = "gray")

Shape palette from Tremmel (1995) (discrete)

Description

Based on experiments Tremmel (1995) suggests the following shape palettes:

Usage

tremmel_shape_pal(overlap = FALSE, alt = FALSE)

Arguments

Details

If two symbols, then use a solid circle and plus sign.

If three symbols, then use a solid circle, empty circle, and an empty triangle. However, that set of symbols does not satisfy the requirement that each symbol should differ from the other symbols in the same feature dimension. A set of three symbols that satisfies this is a circle (curvature), plus sign (number of terminators), triangle (line orientation).

This palette supports up to three values. If more than three groups of data, then separate the groups into different plots.

Value

A palette function. It takes the number of shapes n and returns an integer vector of n shape (pch) codes, and can be used as the palette argument of discrete_scale().

References

Tremmel, Lothar, (1995) "The Visual Separability of Plotting Symbols in Scatterplots" Journal of Computational and Graphical Statistics, https://www.jstor.org/stable/1390760

See Also

Other shapes: circlefill_shape_pal(), cleveland_shape_pal(), scale_shape_circlefill(), scale_shape_cleveland(), scale_shape_tremmel()

Examples

tremmel_shape_pal()(3)

# Alternative triple: solid circle, plus sign and empty triangle
tremmel_shape_pal(alt = TRUE)(3)

show_shapes(tremmel_shape_pal()(3))

Wall Street Journal color palette (discrete)

Description

The Wall Street Journal uses many different color palettes in its plots. This collects a few of them, but is by no means exhaustive. Collections of these plots can be found on the WSJ Graphics X (formerly Twitter) feed and Pinterest.

Usage

wsj_pal(palette = "colors6")

Arguments

Value

A palette function. It takes the number of colours n and returns a character vector of n hex colours, and can be used as the palette argument of discrete_scale().

Palettes

The following palettes are defined,

rgby

Red/Green/Blue/Yellow theme.

red_green

Green/red two-color scale for good/bad.

green_black

Black-green 4-color scale for 'Very negative', 'Somewhat negative', 'somewhat positive', 'very positive'.

dem_rep

Democrat/Republican/Undecided blue/red/gray scale.

colors6

Red, blue, gold, green, orange, and black palette.

See Also

Other colour wsj: scale_colour_wsj()

Examples

wsj_pal()(6)

scales::show_col(wsj_pal("rgby")(4))