Skip to content
Pytorch implementation of Quaternion Knowledge Graph Embedding.
Python C C++ Objective-C Shell
Branch: master
Clone or download
Latest commit 7cba221 Aug 11, 2019
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
base first commit Apr 6, 2019
benchmarks first commit Apr 6, 2019
config first commit Apr 6, 2019
models Update QuatE.py May 2, 2019
release first commit Apr 6, 2019
README.md Update README.md Aug 11, 2019
make.sh first commit Apr 6, 2019
train_OctonionE.py add octonionE Apr 26, 2019
train_QuatE_FB15K.py Update train_QuatE_FB15K.py May 7, 2019
train_QuatE_FB15K237.py Update train_QuatE_FB15K237.py May 7, 2019
train_QuatE_WN18.py Update train_QuatE_WN18.py Apr 9, 2019
train_QuatE_WN18RR.py Update train_QuatE_WN18RR.py May 7, 2019

README.md

Implementation of paper "Quaternion Knowledge Graph Embedding"

Hyper-parameters for reproducing the reported results are provided in the train_QuatE_dataset.py.

How to run

Requirements: Pytorch 1.4+

STEP:

  1. export LD_LIBRARY_PATH=/usr/local/lib:/usr/lib:/usr/local/lib64:/usr/lib64
  2. CUDA_VISIBLE_DEVICES=0 python3 train_QuatE_dataset.py

Citation

@article{zhang2019quaternion,
  title={Quaternion Knowledge Graph Embedding},
  author={Zhang, Shuai and Tay, Yi and Yao, Lina and Liu, Qi},
  journal={arXiv preprint arXiv:1904.10281},
  year={2019}
}

This code is based on the OpenKE project.

You can’t perform that action at this time.