A collection of tools to help with binning.
The Binning
is powerful dict which contains many bin related quantities and
is soley constructed from the edges of a binning.
import numpy as np
import binning_utils
binning_utils.Binning(bin_edges=np.linspace(0, 1, 9))
{'num': 4,
'edges': array([0. , 0.25, 0.5 , 0.75, 1. ]),
'centers': array([0.125, 0.375, 0.625, 0.875]),
'widths': array([0.25, 0.25, 0.25, 0.25]),
'start': 0.0,
'stop': 1.0,
'limits': array([0., 1.])}
Draw the bin to which a sample is randomly assigned to. This assumes that the distribution of samples is uniform over the entire range of the binning. You can provide the aperture of the bins though. (This is the widths of the bins if the binning is one dimensional).
import numpy as np
import binning_utils
prng = np.random.Generator(np.random.PCG64(19))
assignment = binning_utils.draw_random_bin(
prng=prng,
bin_apertures=[1000, 4000, 2000, 3000],
size=10000,
)
print(np.unique(assignment, return_counts=True))
(array([0, 1, 2, 3]), array([ 974, 3950, 2060, 3016]))
Create binning in geomspace which is aligned to decades.
import binning_utils
binning_utils.power10.space(
start_decade=0,
start_bin=0,
stop_decade=2,
stop_bin=1,
num_bins_per_decade=3,
)
array([ 1., 2.15, 4.64, 10., 21.54, 46.41, 100.])
To make bin edges for distributions occuring in power laws.
For example to histogram the energies of cosmic rays which occur in a
power law with slope -2.7
import binning_utils
binning_utils.powerspace(
start=1,
stop=10,
power_slope=-2.7,
size=10,
)
array([ 1. , 1.07017144, 1.15544801, 1.26196439, 1.39995703,
1.58808152, 1.86493297, 2.32807878, 3.33799855, 10. ])
Make tiles of roughly same areas on the surface of a sphere using a Fibonacci spacing.
import binning_utils
vertices_on_sphere = binning_utils.sphere.fibonacci_space(
size=100,
max_zenith_distance_rad=0.5,
)