TorchKGE is an open-source Python library based on PyTorch that aims at implementing in a unified way all existing methods for knowledge-graph embedding. Currently, are implemented : TransE, TransH, TransR, TransD, RESCAL, DistMult, HolE, ComplEx, Analogy. More models should be coming soon (e.g. deep ones such as RGCN, ConvE, ConvKB).

I’d like it to be a clean library that respects PyTorch idiom. Feel free to contribute directly in the repo or just by sending your remarks.

Documentation: https://torchkge.readthedocs.io/en/latest/?badge=latest

PyPi: https://pypi.org/project/torchkge/

Github repo: https://github.com/torchkge-team/torchkge