An implementation of the ComplEx model for Link Prediction in Knowledge Graphs.
The ComplEx model was proposed in ICML 2016 by Theo Trouillon, Johannes Welbl, Sebastian Riedel, Eric Gaussier and Guillaume Bouchard in their ICML 2016 paper Complex Embeddings for Simple Link Prediction.
This implementation is heavily based on the state-of-the-art implementation provided by Timothée Lacroix, Nicolas Usunier and Guillaume Obozinski for their ICML 2018 paper Canonical Tensor Decomposition for Knowledge Base Completion.