when I read your paper, Multi-hop Knowledge Graph Reasoning Based on Hyperbolic Knowledge Graph Embedding and Reinforcement Learning, it seems to be possible to get KG Embedding in Hyperbolic Space in the part, section 4.3. Hyperbolic Knowledge Graph Embedding from what I understand.
I wonder how can I get the hyperbolic knowledge graph embedding when I run ./experiment-emb.sh configs/-conve.sh --train .
From what I've seen and understood the code, it seems to reflect the KG Embedding obtained by training convolution layer in Euclidean space.
Thank you for sharing the code.
when I read your paper, Multi-hop Knowledge Graph Reasoning Based on Hyperbolic Knowledge Graph Embedding and Reinforcement Learning, it seems to be possible to get KG Embedding in Hyperbolic Space in the part, section 4.3. Hyperbolic Knowledge Graph Embedding from what I understand.
I wonder how can I get the hyperbolic knowledge graph embedding when I run ./experiment-emb.sh configs/-conve.sh --train .
From what I've seen and understood the code, it seems to reflect the KG Embedding obtained by training convolution layer in Euclidean space.
Thank you for sharing the code.