-
https://github.com/Coder-Yu/SELFRec/blob/d18d78aa8a1946a26d326ec408edbf560848a1f6/model/graph/LightGCN.py#L26
这里的l2正则为什么还需要除以batchsize?
-
我将原文代码`usrEmbeds, itmEmbeds = self.model(encoderAdj, decoderAdj)`改成`usrEmbeds, itmEmbeds = self.model(self.handler.torchBiAdj)`来模拟lightgcn模型,其他地方不动,最后的模型结果recall@20来到了0.079远高于原文中提到的0.0761
-
你好,我对这篇论文比较感兴趣,然后拉取代码跑了一下,因为代码有问题,只跑了一个yelp2018数据集上的w/o-AU消融实验。
但是发现ProtoAU-w/o-AU recall@20(自己跑出来的)甚至不到LightGCN recall@20(论文中提供的数据)的一半,ProtoAU-w/o-AU recall@20仅有0.06
![image](https://github.com/oce…
-
Hello, I am very interested in reading your article, but there are some issues in reproducing your experiment. The part mentioned in the article that combines with the large model in RLMRec-Con and RL…
-
大佬你好,我是RS这块的菜鸟,我发现按照默认设置运行的话好像没有使用到验证集,验证集是必要的吗?
-
### Description
```
2024-02-19T18:34:57.2553239Z @pytest.mark.gpu
2024-02-19T18:34:57.2568702Z def test_model_lightgcn(deeprec_resource_path, deeprec_config_path):
2024-02-19T18:34:57.…
-
首先感謝你付出,這是一個很實用且容易理解的框架。
除了Qrec, 我也一直有使用別人的代碼或者框架。但一直不懂爲什麽LightGCN在這裏或者SelfRec能有這麽好的效果(precision@10≈15.8),但是在別的框架和原作者代碼就很低(precision@10≈10.1)。請問你們有對比過LightGCN原代碼,找到原因嗎?貌似除了LightGCN,很多別的模型也是。
期待答…
-
### Description
I could not find easily which models support GPU (I had to search in the documentation page with ctrl+F). To check which models support GPU programmatically, one can run (not all mode…
lgabs updated
7 months ago
-
Don't DICE, MACR, PDA have different implementations of MF and LightGCN? Why is it not reflected in this code
-
# Background
- 아이디어 교환, 정리를 위하여 만든 이슈입니다.
- 업데이트 되거나 삭제될 수 있습니다.
# How to FE
## 만든 feature
elapsed
- 유저 개인이 문제를 푼 시간을 sequential하게 나열하여 문제를 푼 시간 정의
- 이상치를 제거하고 Tag에 따른 문제푸는 평균 시간을 바탕으로 재정의
…