wenjiewwj / denoisingrec Goto Github PK
View Code? Open in Web Editor NEWAdaptive Denoising Training (ADT) for Recommendation.
Adaptive Denoising Training (ADT) for Recommendation.
Hi, thanks for your great work. I'm very interested in your paper.
Could you please provide the dataset preprocessing codes for generating the uploaded data? If not, could you give some step-by-step descriptions?
Thanks for any possible help.
作者您好。我运行了你开源的代码,有一些疑惑想要咨询一下您。
1、我在Yelp数据集上运行了GMF+T-CE模型,具体的参数设置是:batch_size=1024, dataset='yelp', drop_rate=0.2, dropout=0.0, epochs=10, eval_freq=2000, exponent=1, factor_num=32, gpu='3', lr=0.001, model='GMF', num_gradual=30000, num_layers=3, num_ng=1, out=True, top_k=[50, 100],这些都是默认的设置,我仅仅改变了训练次数epoch,即epoch =[10, 20, 30],训练10个epoch时,Recall=[0.0875, 0.1452], NDCG=[0.0357,0.0487]; 训练20个epoch时,Recall=[0.0941,0.1550], NDCG=[0.0382,0.0520];训练30个epoch时,Recall=[0.1021,0.1649], NDCG=[0.0420,0.0562]。可以看到性能还是一直在大幅上升状态,即训练10个epoch并没有让模型达到收敛状态。我观察了论文中实验结果表3,其中的实验结果与epoch=10时的基本一致,这样的对比是否有失公允呢?
2、虽然代码中设置了固定的随机数种子,但每次实验结果似乎都不同,并且和论文中的结果对不上,是因为多次取平均的结果吗?
非常感谢您在百忙之中阅读我的问题,期待您的回复。 @WenjieWWJ
Hi, when I perform "python inference.py --dataset=yelp --model=GMF --drop_rate=0.1 --num_gradual=30000 --gpu=0", I obtain Recall@50 and Recall@100 both at 0.0000 though NDCG is the same as that of the paper. Is there something wrong in your source code?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.