Movatterモバイル変換


[0]ホーム

URL:


Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

Sign up
Appearance settings

Author: Tong Zhao (tzhao2@nd.edu). ICML 2022. Learning from Counterfactual Links for Link Prediction

License

NotificationsYou must be signed in to change notification settings

DM2-ND/CFLP

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

This repository contains the source code for the ICML 2022 paper:

Learning from Counterfactual Links for Link Prediction

byTong Zhao (tzhao2@nd.edu),Gang Liu,Daheng Wang,Wenhao Yu, andMeng Jiang.

Requirements

This code package was developed and tested with Python 3.8.5, PyTorch 1.6.0, and PyG 1.6.1. All dependencies specified in therequirements.txt file. The packages can be installed by

pip install -r requirements.txt

Usage

The step of finding all the counterfactual links can be slow for the first run, please adjust the--n_workers parameter according to available processes if you are trying out different settings. The cached files for the counterfactual links that were used in our experiments can be foundhere, please download and put them underdata/T_files/ before reproducing our experiments.

Following are the commands to reproduce our experiment results on different datasets.

# Corapython main.py --dataset cora --metric auc --alpha 1 --beta 1 --gamma 30 --lr 0.1 --embraw mvgrl --t kcore --neg_rate 50 --jk_mode mean --trail 20# CiteSeerpython main.py --dataset citeseer --metric auc --alpha 1 --beta 1 --gamma 30 --lr=0.1 --embraw dgi --t kcore --neg_rate 50 --jk_mode mean --trail 20# PubMedpython main.py --dataset pubmed --metric auc --alpha 1 --beta 1 --gamma 30 --lr 0.1 --embraw mvgrl --t kcore --neg_rate 40 --jk_mode mean --batch_size 12000 --epochs 200 --patience 50 --trail 20# Facebookpython main.py --dataset facebook --metric hits@20 --alpha 1e-3 --beta 1e-3 --gamma 30 --lr 0.005 --embraw mvgrl --t louvain --neg_rate 1 --jk_mode mean --trail 20# OGBL-ddipython main.py --dataset ogbl-ddi --metric hits@20 --alpha 1e-3 --beta 1e-3 --gamma 10 --lr 0.01 --embraw dgi --t louvain  --neg_rate 1 --jk_mode mean --epochs=200 --epochs_ft=200 --patience=50 --trail 20

Cite

If you find this repository useful in your research, please cite our paper:

@inproceedings{zhao2022learning,title={Learning from Counterfactual Links for Link Prediction},author={Zhao, Tong and Liu, Gang and Wang, Daheng and Yu, Wenhao and Jiang, Meng},booktitle={International Conference on Machine Learning},pages={26911--26926},year={2022},organization={PMLR}}

About

Author: Tong Zhao (tzhao2@nd.edu). ICML 2022. Learning from Counterfactual Links for Link Prediction

Topics

Resources

License

Stars

Watchers

Forks

Languages


[8]ページ先頭

©2009-2026 Movatter.jp