Fidelity calculation requires '_edge_mask' as parameter

[SvenWeinzierl]

🐛 Describe the bug

Hi there,

I tried to calculate fidelity for the GNNExplainer example (https://github.com/pyg-team/pytorch_geometric/blob/master/examples/gnn_explainer.py). In doing that, I got the following error:

raise TypeError("cannot assign '{}' as parameter '{}' " TypeError: cannot assign 'torch.FloatTensor' as parameter '_edge_mask' (torch.nn.Parameter or None expected)

Code:

import os.path as osp

import torch
import torch.nn.functional as F

from torch_geometric.datasets import Planetoid
from torch_geometric.explain import Explainer, GNNExplainer, metric
from torch_geometric.nn import GCNConv

dataset = 'Cora'
path = osp.join(osp.dirname(osp.realpath(__file__)), '..', 'data', 'Planetoid')
dataset = Planetoid(path, dataset)
data = dataset[0]

class Net(torch.nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = GCNConv(dataset.num_features, 16)
        self.conv2 = GCNConv(16, dataset.num_classes)

    def forward(self, x, edge_index):
        x = F.relu(self.conv1(x, edge_index))
        x = F.dropout(x, training=self.training)
        x = self.conv2(x, edge_index)
        return F.log_softmax(x, dim=1)

device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
model = Net().to(device)
data = data.to(device)
optimizer = torch.optim.Adam(model.parameters(), lr=0.01, weight_decay=5e-4)

for epoch in range(1, 10):
    model.train()
    optimizer.zero_grad()
    out = model(data.x, data.edge_index)
    loss = F.nll_loss(out[data.train_mask], data.y[data.train_mask])
    loss.backward()
    optimizer.step()

explainer = Explainer(
    model=model,
    algorithm=GNNExplainer(epochs=100),
    explanation_type='model',
    node_mask_type='attributes',
    edge_mask_type='object',
    model_config=dict(
        mode='multiclass_classification',
        task_level='node',
        return_type='log_probs',
    ),
)
node_index = 10
explanation = explainer(data.x, data.edge_index, index=node_index)
print(f'Generated explanations in {explanation.available_explanations}')

print(metric.fidelity(explainer, explanation))

Environment

• PyG version:
• PyTorch version:
• OS:
• Python version:
• CUDA/cuDNN version:
• How you installed PyTorch and PyG (conda, pip, source):
• Any other relevant information (e.g., version of torch-scatter):

[rusty1s]

Thanks for reporting. Will be fixed in https://github.com/pyg-team/pytorch_geometric/pull/6510.