Official code implementation for SIGIR 23 paper Normalizing Flow-based Neural Process for Few-Shot Knowledge Graph Completion
pytorch==1.11
tqdm==4.64
normflows==1.4
dgl==0.9.0
tensorboardx==2.5.1
Note: Please make sure dgl==0.9.0 and use CUDA, our codes rely on a small bug of dgl for running.
- python 3.8
- Ubuntu 22.04
- RTX3090/A100
- Memory 32G/128G
Download the datasets and extract to the project root folder.
NELL (3090)
python main.py --dataset NELL-One --data_path ./NELL --few 5 --data_form Pre-Train --prefix np_rgcn_attn_planar_nellone_5shot_intrain --device 0 --batch_size 128 --flow Planar --g_batch 1024WIKI (A100)
python main.py --dataset Wiki-One --data_path ./Wiki --few 5 --data_form Pre-Train --prefix np_rgcn_attn_planar_wiki_5shot_intrain_g_batch_1024_eval_8 --device 0 --batch_size 64 --flow Planar -dim 50 --g_batch 1024 --eval_batch 8 --eval_epoch 4000FB15K-237 (3090)
python main.py --dataset FB15K-One --data_path ./FB15K --few 5 --data_form Pre-Train --prefix np_rgcn_attn_planar_fb15k_5shot_intrain --device 0 --batch_size 128 --flow Planar --g_batch 1024 --eval_batch_size 128 --K 14Download the checkpoint and extract to the state/ folder.
NELL
python main.py --dataset NELL-One --data_path ./NELL --few 5 --data_form Pre-Train --prefix np_rgcn_attn_planar_nellone_5shot_intrain_0.46 --device 0 --batch_size 128 --flow Planar --g_batch 1024 --step testWIKI
python main.py --dataset Wiki-One --data_path ./Wiki --few 5 --data_form Pre-Train --prefix np_rgcn_attn_planar_wiki_5shot_intrain_g_batch_1024_eval_8_0.503 --device 0 --batch_size 64 --flow Planar -dim 50 --g_batch 1024 --eval_batch 8 --eval_epoch 4000 --step testFB15K-237
python main.py --dataset FB15K-One --data_path ./FB15K --few 5 --data_form Pre-Train --prefix np_rgcn_attn_planar_fb15k_5shot_intrain_0.536 --device 0 --batch_size 128 --flow Planar --g_batch 1024 --eval_batch_size 128 --K 14 --step test5-shot FKGC results
| Dataset | MRR | Hits@10 | Hits@5 | Hits@1 |
|---|---|---|---|---|
| NELL | 0.460 | 0.494 | 0.471 | 0.437 |
| WIKI | 0.503 | 0.668 | 0.599 | 0.423 |
| FB15K-237 | 0.538 | 0.671 | 0.593 | 0.476 |
See full results in our paper.
If you use this repo, please cite the following paper.
@inproceedings{luo2023npfkgc,
title = {Normalizing Flow-based Neural Process for Few-Shot Knowledge Graph Completion},
author = {Luo, Linhao and Li, Yuan-Fang and Haffari, Gholamreza and Pan, Shirui},
booktitle = {The 46th International ACM SIGIR Conference on Research and Development in Information Retrieval},
year = {2023}
}
This repo is mainly based on GANA. We thank the authors for their great works.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent