Social circles

This example inspired by SocNetV (https://socnetv.org/data/uploads/screenshots/32/socnetv-32-example-dot.png), shows social connections within circles. It also demonstrates how to combine iplotx with other matplotlib artists.

import igraph as ig
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import iplotx as ipx


edge_data = [
    (1, 2),
    (2, 3),
    (3, 4),
    (4, 5),
    (5, 1),
    (1, 6),
    (6, 7),
    (7, 8),
    (8, 9),
    (9, 10),
]
edge_data = pd.DataFrame(edge_data).astype(str)

g = ig.Graph.DataFrame(
    edge_data,
    use_vids=False,
    directed=True,
)


def coords(radius, angle):
    angle = np.deg2rad(angle)
    return radius * np.cos(angle), radius * np.sin(angle)


radii = [1, 3, 5, 6, 7, 12]
layout = [
    coords(radii[0], 0),
    coords(radii[4], -30),
    coords(radii[3], -70),
    coords(radii[2], -100),
    coords(radii[1], -135),
    coords(radii[5], 180),
    coords(radii[5], 180 - 36),
    coords(radii[5], 180 - 36 * 2),
    coords(radii[5], 180 - 36 * 3),
    coords(radii[5], 180 - 36 * 4),
]


fig, ax = plt.subplots(figsize=(8, 8))
ipx.network(
    g,
    layout,
    ax=ax,
    vertex_marker="s",
    vertex_facecolor="red",
    edge_shrink=3,
    edge_zorder=2,
    vertex_zorder=3,
    vertex_label_color="black",
    vertex_labels=list(g.vs["name"]),
    zorder=3,
)

for radius in radii:
    ax.add_patch(
        plt.Circle(
            (0, 0), radius,
            facecolor="none", edgecolor="red", ls="dashed",
            zorder=1,
        )
    )

ax.set(xlim=(-13, 13), ylim=(-13, 13))
fig.tight_layout()