Code for "CompoundE: Knowledge Graph Embedding with Translation, Rotation and Scaling Compound Operations"
https://github.com/snap-stanford/ogb
- Create a python 3.8.8 environment using Anaconda
- Install torch 1.11.0 using Anaconda
- Install torch-geometric using Anaconda
- Install ogb using pip
- Install other packages using pip
CUDA_VISIBLE_DEVICES=0 python run.py --do_train --cuda --do_valid --do_test --evaluate_train \
--model CompoundE -n 250 -b 4096 -d 100 -g 7 -a 1.0 -adv -tr \
-lr 0.005 --max_steps 300000 --cpu_num 8 --test_batch_size 32 --print_on_screen
We refer to the code of PairRE. Thanks for their contributions.
Citation
If you find the source codes useful, please consider citing our paper:
@inproceedings{ge2023compounding,
title={Compounding geometric operations for knowledge graph completion},
author={Ge, Xiou and Wang, Yun Cheng and Wang, Bin and Kuo, C-C Jay},
booktitle={Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
pages={6947--6965},
year={2023}
}