aaanalysis.plot_get_cmap

aaanalysis.plot_get_cmap(name='CPP', n_colors=101, facecolor_dark=False)

Get colormaps specified for AAanalysis.

Added in version 0.1.2.

Parameters:

• name ({'CPP', 'SHAP'}, default='CPP') –

  The name of the AAanalysis color palettes.

  • CPP: Continuous colormap for CPP plots.

  • SHAP: Continuous colormap for CPP-SHP plots.

• n_colors (int, default=101) – Number of colors. Must be at least 3.

• facecolor_dark (bool, optional) – Whether central color in is black (if True) or white (if False).

Returns:

List with colors given as RGB tuples.

Return type:

cmap

See also

• Plotting Prelude.

• Matplotlib color names

• seaborn.color_palette() function to generate a color palette in seaborn.

• seaborn.light_palette function() to generate a lighter color palettes.

• The SHAP package.

Examples

Continuous color maps (cmap) for ‘CPP plots’ and ‘CPP-SHAP plots’ can be retrieved using aa.plot_get_cmap(), where the number of colors is set using n_colors:

import matplotlib.pyplot as plt
import seaborn as sns
import aaanalysis as aa
data = {'Classes': ['Negative values', 'Middle value (0)', 'Positive values',], 'Values': [13, 23, 33]}
aa.plot_settings(font_scale=0.9)
colors = aa.plot_get_cmap(n_colors=3)
sns.barplot(data=data, x='Classes', y='Values', palette=colors, hue="Classes")
plt.show()

For ‘CPP plots’, we recommend using a white facecolor using facecolor_dark=False:

colors = aa.plot_get_cmap(name="CPP", n_colors=3, facecolor_dark=False)
sns.barplot(data=data, x='Classes', y='Values', palette=colors, hue="Classes",
            edgecolor="black")
plt.show()

For ‘CPP-SHAP plots’, we recommend using a dark facecolor:

colors = aa.plot_get_cmap(name="SHAP", n_colors=3, facecolor_dark=True)
sns.barplot(data=data, x='Classes', y='Values', palette=colors, hue="Classes")
plt.show()

The number of colors steps can be adjusted to any number integer number:

n = 11
colors = aa.plot_get_cmap(n_colors=n)
sns.palplot(colors)
plt.show()
colors = aa.plot_get_cmap(n_colors=n, facecolor_dark=False)
sns.palplot(colors)
plt.show()
colors = aa.plot_get_cmap(n_colors=n, facecolor_dark=True)
sns.palplot(colors)
plt.show()