Giters

abojchevski / sparse_smoothing

Implementation of the certificates proposed in the paper "Efficient Robustness Certificates for Discrete Data: Sparsity-Aware Randomized Smoothing for Graphs, Images and More"

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

problem encountered running demo.py

msqiqi10 opened this issue · comments

commented

Thanks for your contribution!

When I was trying to incorporate your work as a baseline, I'm having this error:
In sparse_smoothing/models.py, line 18 edge_idx, norm = self.norm(edge_idx, x.size(0), None, self.improved, x.dtype)
'SparseGCNConv' object has no attribute 'norm'.
Could you please help me out?

Thanks!

commented

Hi,

In the new version of Pytorch Geometric the way the normalization term for GCN models is computed was changed. I've updated the model to match the new version. Does it work for you now?