I would like to use GNNExplainer to interpret the model. | Sololearn: Learn to code for FREE!
Novo curso! Todo programador deveria aprender IA generativa!
Experimente uma aula grƔtis
0

I would like to use GNNExplainer to interpret the model.

I developed a GCN model using pytorch. (without using geometry) I want to put the developed model into GNNExplainer of geometry and check the result, but an error occurs. The matrices I used to create the GCN are featrue(150,65) and adjacency matrix (150,150). Isn't it possible to use x = feature / edge_index = adjacency matrix for GNNExplainer input?

4th Dec 2020, 6:29 AM
ģ”°ķ˜•ģ‹
ģ”°ķ˜•ģ‹ - avatar
1 Resposta
4th Dec 2020, 7:22 AM
Piyush
Piyush - avatar