moetayuko / HPNC

Hyperspherical Prototype Node Clustering

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Hyperspherical Prototype Node Clustering

This is the official implementation of our TMLR paper Hyperspherical Prototype Node Clustering.

@article{
lu2024hyperspherical,
title={Hyperspherical Prototype Node Clustering},
author={Jitao Lu and Danyang Wu and Feiping Nie and Rong Wang and Xuelong Li},
journal={Transactions on Machine Learning Research},
issn={2835-8856},
year={2024},
url={https://openreview.net/forum?id=z3ZlnaOM0d},
}

Prepare Environment

conda create -n hpnc
conda activate hpnc
conda install pytorch pytorch-cuda=11.8 pyg yacs pytorch-lightning tensorboardx -c pyg -c pytorch -c nvidia

Reproduce the Results

To reproduce the results in Table 1, run with:

./run_all.sh

Then, find the results from results/hpnc_{dec,rim}/$dataset/agg/train/best.json

To reproduce the results in Table 2, run with:

./run_all.sh --ablation

About

Hyperspherical Prototype Node Clustering

License:MIT License


Languages

Language:Python 99.0%Language:Shell 1.0%