KDEGlyph¶

KDEGlyph evaluates an isotropic Gaussian kernel-density estimate of an (x, y) point cloud on a grid (NumPy only, no scipy) and draws it as filled (shade=True, default) or line density contours.

In [1]:

%matplotlib inline
import numpy as np
import matplotlib.pyplot as plt
from cleopatra.kde_glyph import KDEGlyph
rng = np.random.default_rng(0)
x = np.concatenate([rng.normal(-1, 0.6, 400), rng.normal(1.5, 0.5, 300)])
y = np.concatenate([rng.normal(0, 0.6, 400), rng.normal(1.0, 0.5, 300)])

%matplotlib inline import numpy as np import matplotlib.pyplot as plt from cleopatra.kde_glyph import KDEGlyph rng = np.random.default_rng(0) x = np.concatenate([rng.normal(-1, 0.6, 400), rng.normal(1.5, 0.5, 300)]) y = np.concatenate([rng.normal(0, 0.6, 400), rng.normal(1.0, 0.5, 300)])

1. Filled density contours¶

In [2]:

kde = KDEGlyph(x, y, cmap='viridis')
fig, ax, cs = kde.plot(title='Filled KDE')

kde = KDEGlyph(x, y, cmap='viridis') fig, ax, cs = kde.plot(title='Filled KDE')

2. Line contours with a wider kernel¶

In [3]:

kde = KDEGlyph(x, y, shade=False, bw_method=1.4, levels=12, cmap='inferno')
fig, ax, cs = kde.plot(title='Line KDE')

kde = KDEGlyph(x, y, shade=False, bw_method=1.4, levels=12, cmap='inferno') fig, ax, cs = kde.plot(title='Line KDE')