LightGCN-pytorch

This is the Pytorch implementation for our SIGIR 2020 paper:

SIGIR 2020. Xiangnan He, Kuan Deng ,Xiang Wang, Yan Li, Yongdong Zhang, Meng Wang(2020). LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation, Paper in arXiv.

Author: Prof. Xiangnan He (staff.ustc.edu.cn/~hexn/)

(Also see Tensorflow implementation)

Introduction

In this work, we aim to simplify the design of GCN to make it more concise and appropriate for recommendation. We propose a new model named LightGCN,including only the most essential component in GCN—neighborhood aggregation—for collaborative filtering

Enviroment Requirement

pip install -r requirements.txt

Dataset

We provide three processed datasets: Gowalla, Yelp2018 and Amazon-book and one small dataset LastFM.

see more in dataloader.py

An example to run a 3-layer LightGCN

run LightGCN on Gowalla dataset:

• command

cd code && python main.py --decay=1e-4 --lr=0.001 --layer=3 --seed=2020 --dataset="gowalla" --topks=[20] --recdim=64

• log output

...
======================
EPOCH[5/1000]
BPR[sample time][16.2=15.84+0.42]
[saved][[BPR[aver loss1.128e-01]]
[0;30;43m[TEST][0m
{'precision': array([0.03315359]), 'recall': array([0.10711388]), 'ndcg': array([0.08940792])}
[TOTAL TIME] 35.9975962638855
...
======================
EPOCH[116/1000]
BPR[sample time][16.9=16.60+0.45]
[saved][[BPR[aver loss2.056e-02]]
[TOTAL TIME] 30.99874997138977
...

NOTE:

1. Even though we offer the code to split user-item matrix for matrix multiplication, we strongly suggest you don't enable it since it will extremely slow down the training speed.
2. If you feel the test process is slow, try to increase the testbatch and enable multicore(Windows system may encounter problems with multicore option enabled)
3. Use tensorboard option, it's good.
4. Since we fix the seed(--seed=2020 ) of numpy and torch in the beginning, if you run the command as we do above, you should have the exact output log despite the running time (check your output of epoch 5 and epoch 116).

Extend:

• If you want to run lightGCN on your own dataset, you should go to dataloader.py, and implement a dataloader inherited from BasicDataset. Then register it in register.py.
• If you want to run your own models on the datasets we offer, you should go to model.py, and implement a model inherited from BasicModel. Then register it in register.py.
• If you want to run your own sampling methods on the datasets and models we offer, you should go to Procedure.py, and implement a function. Then modify the corresponding code in main.py

Results

all metrics is under top-20

tensorflow version results:

pytorch version results (stop at 1000 epochs):

(for seed=2020)

• gowalla:

	Recall	ndcg	precision
layer=1	0.1687	0.1417	0.05106
layer=2	0.1786	0.1524	0.05456
layer=3	0.1824	0.1547	0.05589
layer=4	0.1825	0.1537	0.05576

NOTE: layers=4 we use seed=1000 to attain a better performance

• yelp2018

	Recall	ndcg	precision
layer=1	0.05604	0.4557	0.02519
layer=2	0.05988	0.04956	0.0271
layer=3	0.06347	0.05238	0.0285
layer=4	0.06515	0.05325	0.02917

For those who want the well-trained models, please e-mail me ( gusye AT mail.ustc.edu.cn)