relation-gcn-pytorch

Pytorch implementation of 'Modeling relational data with graph convolutional networks', ESWC, 2018.

Dependencies

• pytorch 1.1.0
• numpy 1.16.4
• scipy 1.3.0

Results

AIFB

Epoch 0: train loss 1.34 val loss 1.31 val acc0.42 Epoch 10: train loss 0.12 val loss 0.2 val acc0.94 Epoch 20: train loss 0.0 val loss 0.35 val acc0.86 Epoch 30: train loss 0.0 val loss 0.51 val acc0.89 Epoch 40: train loss 0.0 val loss 0.51 val acc0.89

References

[1] Schlichtkrull M, Kipf T N, Bloem P, et al. Modeling relational data with graph convolutional networks[C]//European Semantic Web Conference. Springer, Cham, 2018: 593-607.