Comparing radius samplers

Radius samplers are explicit physical inputs to an RSA simulation. Their to_bins method converts a continuous distribution into radius classes and number fractions for use by the analytical Percus–Yevick model.

import matplotlib.pyplot as plt
import numpy as np

from PackLab import samplers
from PackLab.units import ureg


uniform = samplers.UniformRadiusSampler(
    minimum_radius=80 * ureg.nanometer,
    maximum_radius=160 * ureg.nanometer,
    bins=8,
)
log_normal = samplers.LogNormalRadiusSampler(
    median_radius=110 * ureg.nanometer,
    geometric_standard_deviation=1.25,
    maximum_radius_clip=250 * ureg.nanometer,
    bins=8,
)

figure, axes = plt.subplots(1, 2, figsize=(9, 3.5), sharey=True)
for axis, sampler, label in zip(axes, (uniform, log_normal), ("Uniform", "Log-normal")):
    _ = sampler.plot_histogram(ax=axis, color="tab:blue", width=7)
    axis.set_title(label)
figure.tight_layout()

assert np.isclose(sum(uniform.to_bins()[1]), 1.0)
assert np.isclose(sum(log_normal.to_bins()[1]), 1.0)