Numbering of node and feature of node in the graph

[kimsijin33]

I wondering of feature of node how to learning in GCN model.

if there are below data.
import matplotlib.pyplot as plt
import networkx as nx

nodes is 0, 1, 2, 3, 4.
Total there are node of 5.

Are below both data equal rigth ?
graph[98].nodes()
tensor([0,…
graph[98].ndata[‘x’]
tensor([66,…
so,
nodes = ndata[‘x’]
0 = 66
1 = 20
2 = 144
3 = 2254
4 = 2712
above content right?

Or like these rigth?
graph[98].ndata[‘0’] = tensor([66])
graph[98].ndata[‘1’] = tensor([20])
graph[98].ndata[‘2’] = tensor([144])
graph[98].ndata[‘3’] = tensor([2254])
graph[98].ndata[‘4’] = tensor([2712])

I think no learning at GCN model features of nodes now like below not at all.
graph[98].ndata[‘x’] = tensor([ 66, 20, 144, 2254, 2712])

What are learing in GCN model?

Thank you.

[minjie]

Hi, I’m not sure what is the problem you’ve encountered. Are you asking how to set node features to a graph?

[kimsijin33]

Yeah, so could I learning feature of node in GCN model?

or How to learning feature of node in GCN model?