In Search of the Perfect Palette

Three types of palettes:

1. discriminate

2. qualitative

3. smooth

The {ggplot2} default colors

ggplot(mpg, aes(displ, hwy)) +
  geom_point(aes(color = class)) +
  geom_smooth()

-> not color blindness friendly

Quick switch to Okabe-Ito

ggplot(mpg, aes(displ, hwy)) +
  geom_point(aes(color = class)) +
  geom_smooth() +
  scale_colour_manual(values = palette.colors(palette = "Okabe-Ito"))

The Nature standard in academic publishing. See Wong (2011).

The same in Python

import matplotlib.pyplot as plt

# Okabe-Ito as the default colour cycle (built into matplotlib 3.11+)
okabe_ito = plt.colormaps["okabe_ito"].colors
plt.rcParams["axes.prop_cycle"] = plt.cycler(color=okabe_ito)

import pandas as pd

mpg = pd.read_csv(
    "https://raw.githubusercontent.com/tidyverse/ggplot2/main/data-raw/mpg.csv"
)

for cls, d in mpg.groupby("class"):
    plt.scatter(d.displ, d.hwy, label=cls)
plt.legend()
plt.show()