如何在igraph网络图中使用权重列表设置线条宽度

The issue you're running into is that Plotly's Scatter3d trace doesn't support assigning multiple line widths in a single trace—the line.width parameter expects a single numeric value, not a list. To get edge-specific line widths, you need to create a separate Scatter3d trace for each edge, each using its own weight as the line width.

Here's how to adjust your code to make this work:

First, generate a 3D layout for your graph (if you haven't already done this to get Xe/Ye/Ze):

import igraph as ig
import plotly.graph_objects as go

# Your original graph setup
eG = ig.Graph(directed=False)
eG.add_vertices(6)
eG.add_edges([(0,1),(1,2),(2,3),(3,4),(3,5),(5,3)])
eG.es['weight'] = [12,1,2,3,4,1]

# Generate 3D layout (using Kamada-Kawai algorithm for balanced spacing)
layout = eG.layout('kk_3d')
vertex_coords = layout.coords  # List of (x,y,z) tuples for each vertex

Next, create a separate trace for each edge, using the edge's weight to set the line width:

traces = []

# Loop through each edge in the graph
for edge in eG.es:
    source_idx = edge.source
    target_idx = edge.target
    weight = edge['weight']
    
    # Get coordinates for source and target vertices
    x = [vertex_coords[source_idx][0], vertex_coords[target_idx][0]]
    y = [vertex_coords[source_idx][1], vertex_coords[target_idx][1]]
    z = [vertex_coords[source_idx][2], vertex_coords[target_idx][2]]
    
    # Create trace for this edge
    trace = go.Scatter3d(
        x=x, y=y, z=z,
        mode='lines',
        line=dict(color='rgb(125,125,125)', width=weight),
        hoverinfo='none'
    )
    traces.append(trace)

# Add vertex markers (optional but improves readability)
vertex_trace = go.Scatter3d(
    x=[coord[0] for coord in vertex_coords],
    y=[coord[1] for coord in vertex_coords],
    z=[coord[2] for coord in vertex_coords],
    mode='markers',
    marker=dict(size=8, color='rgb(255,0,0)'),
    hoverinfo='text'
)
traces.append(vertex_trace)

# Create and display the figure
fig = go.Figure(data=traces)
fig.update_layout(scene=dict(xaxis_title='X', yaxis_title='Y', zaxis_title='Z'))
fig.show()

Key Changes: #

• Separate traces per edge: Each edge gets its own Scatter3d trace, allowing unique line widths for every connection.
• Direct coordinate mapping: We use the layout coordinates directly to define each edge's start and end points, eliminating the need for concatenated Xe/Ye/Ze lists with NaNs.
• Optional vertex markers: Adding colored markers for vertices makes the network structure easier to interpret, but you can remove this block if it's not needed.

This approach will correctly render each edge with a line width matching its weight—for example, the edge between vertices 0 and 1 (weight 12) will be significantly thicker than edges with weight 1.

内容的提问来源于stack exchange，提问作者Arun Taneja