-
Unbelievably! There is a new bug in running GCMC and GCN model implemented by LightGCN.py. That is, please specify the parameter "--adj_type" to 'norm' or 'mean' (the default is 'pre') . Otherwise, …
-
I do not see any code on how you generated the item embeddings for your datasets. How were the item embeddings created for the Autoencoders? Thanks.
-
### 😎 Responsibility
@wh4044, @juk1329 , @Glanceyes
### 💡Issue
- @juk1329님이 실험한 LightGCN으로 문제 추천 결과 Inference 시 유저의 성향에 과도하게 편향된 결과가 나온다는 보고가 있었습니다.
- LightGCN의 학습된 유저 또는 태그 임베딩을 Multi-D…
-
The results in the original paper (LightGCN) 0.1817 0.2518 0.0655 0.1170 0.0416 0.0739
but the result with the code is Best Iter=[45]@[28563.4] recall=[0.18198], precision=[0.05613], ndcg=[0.15523]
…
LBJ66 updated
3 years ago
-
When setting --pretrain as 1 (loading the trained model such as 'lgn-gowalla-3-64.pth.tar'), the initial of lgn model faces the problem:
Traceback (most recent call last):
File "main.py", line 18,…
-
Dear Author:
Hi, RecBole is used as an integrated recomendation system framework that encapsulates the underlying operations to make rapid development possible. However, It also result in some dif…
-
## Why
推薦・機械学習勉強会は、推薦や機械学習、その周辺技術を通じてサービスを改善することにモチベーションのある人達の集まりです。ニュースやブログから論文まで、気になったものについてお互い共有しましょう！
発信のため、ここは **public** にしてあります。外部からの参加をご希望の方は松村(https://twitter.com/yu__ya4) まで DM を送るか、Wa…
-
Excuse me, when I run this command:
`python LightGCN.py --dataset gowalla --regs [1e-4] --embed_size 64 --layer_size [64,64,64] --lr 0.001 --batch_size 2048 --epoch 20
`
I got the following result:…
-
Hello, thank you for offering this project for Sign-Aware Recommendation.
I have tried this code, it can achieve the results reported in the paper.
But I have some questions about the code and the…
-
作者你好，我看文章中写的实验采用的早停策略是10代早停，请问是所有实验包括对比算法都是用10代早停吗，因为我看lightgcn和ngcf论文中早停设置的是50代停，请问作者在实验中的早停是所有实验都统一设置为10吗？不同的早停代数会不会对结果有影响呢？