spine_score_map¶

The colormap used for SPINE scores, as a plain array.

Pale at score 0 and deep at score 1, monotone in lightness, built from Okabe-Ito anchors so it stays readable in greyscale and to colour-blind readers. The typical bulk recedes and the atypical objects stand out.

spine_data_plot and spine_network_plot apply it for you. Use this directly only when drawing your own figure; it needs numpy alone.

Syntax¶

from entropy_spine import spine_score_map
help(spine_score_map)

Help on function spine_score_map in module entropy_spine.interface.plots:

spine_score_map(n: 'int' = 256) -> 'np.ndarray'
    Sequential colormap for SPINE scores, as an (n, 3) RGB array.

    Pale at score 0 and deep at score 1, monotone in lightness, built from
    Okabe-Ito anchors so it stays readable in greyscale and to colour-blind
    readers. ``spine_data_plot`` and ``spine_network_plot`` apply it for you;
    use this directly only when drawing your own figure.

Using it on your own axes¶

Fix the colour limits at 0 to 1, as the package's own figures do, so two pictures can be compared by eye.

%matplotlib inline
import matplotlib.pyplot as plt
import numpy as np
from matplotlib.colors import ListedColormap
from entropy_spine import spine, spine_score_map

rng = np.random.default_rng(4242)
D = np.vstack([rng.standard_normal((200, 2)),
               [12, 12] + 0.25 * rng.standard_normal((20, 2))])
s, _ = spine(D, "dist2median", "upper")

fig, ax = plt.subplots()
sc = ax.scatter(D[:, 0], D[:, 1], s=30 + 60 * s, c=s,
                cmap=ListedColormap(spine_score_map(256)), vmin=0, vmax=1,
                edgecolors="0.3", linewidths=0.4)
fig.colorbar(sc, label="score")
ax.set_aspect("equal")
ax.set_title("drawn by hand, coloured by score");
No description has been provided for this image

The ramp itself¶

spine_score_map(9).round(3)

array([[0.973, 0.969, 0.882],
       [0.958, 0.946, 0.516],
       [0.941, 0.894, 0.259],
       [0.923, 0.777, 0.084],
       [0.902, 0.624, 0.   ],
       [0.877, 0.493, 0.   ],
       [0.835, 0.369, 0.   ],
       [0.664, 0.255, 0.01 ],
       [0.349, 0.149, 0.031]])

Monotone in lightness¶

Pale at zero and deep at one, so the ordering survives a greyscale print.

lum = spine_score_map(256) @ np.array([0.2126, 0.7152, 0.0722])
print(f"luminance at score 0: {lum[0]:.3f}, at score 1: {lum[-1]:.3f}")

luminance at score 0: 0.964, at score 1: 0.183

See also¶

spine_data_plot, spine_network_plot.