A pure Python implementation of the circle packing algorithm detailed in Wang et al. (2006). Visualization of large hierarchical data by circle packing. Proc of the SIGCHI, 517-520.
Using pip
:
pip install packcircles
or directly using setup.py
:
git clone https://github.com/mhtchan/packcircles.git
cd packcircles
python setup.py install
The function pack
takes an iterable of the radii of the circles to pack and returns a generator that yields the layout of each circle as tuples in the form (x_coordinate, y_coordinate, radius)
.
import packcircles as pc
circles = pc.pack([15,5,7,12])
print(list(circles))
## [(-15.579319782711305, 4.287939798231928, 15), (4.4206802172886945, 4.287939798231928, 5), (2.9206802172886945, -7.6179411015587295, 7), (20.647637933172685, -0.7801804930509242, 12)]
The layout in the gif at the top of the page can be generated by the following code:
import packcircles as pc
import matplotlib.pyplot as plt
from matplotlib.cm import get_cmap
radii = [28,12,51,26,10,16,24,25,59,11,29,40,16,11,10,26,39,16,48,36,28]
fig, ax = plt.subplots()
cmap = get_cmap('coolwarm_r')
circles = pc.pack(radii)
for (x,y,radius) in circles:
patch = plt.Circle(
(x,y),
radius,
color=cmap(radius/max(radii)),
alpha=0.65
)
ax.add_patch(patch)
fig.set_figheight(15)
fig.set_figwidth(15)
ax.set(xlim=(-150, 140), ylim=(-180, 170))
plt.axis('off')