GATNE icon indicating copy to clipboard operation
GATNE copied to clipboard

Source code and dataset for KDD 2019 paper "Representation Learning for Attributed Multiplex Heterogeneous Network"

Results 38 GATNE issues
Sort by recently updated
recently updated
newest added

您好,请问模型现在能处理带权图吗? 期待您的回复!谢谢您

Hi, thanks for providing the code to the public. 1- I've seen your replies on how to use your model for the weighted graph, but I am wondering whether you...

I didnt understand either 1 or 0 denoting an edge or a non-edge? Shouldn't there be an edge between nodes? And what does the non-edge mean?

您好!非常感谢您的开源项目! 我在我的数据集上运行了您的代码,其中的embedding来自TransR训练得到的,其中TransR训练后的Hits@10=0.61,但是对比下来,发现没有使用embedding的效果会比用了的还要好, F1 = 0.744,带了embedding的F1 = 0.689,各项指标都比不过,不知道是什么原因造成的?是TransR得到的embedding没有作用嘛? 打扰了!谢谢!

我分别试验了浏览edge、购买edge、收藏edge一起训练,以及3种edge数据分别训练,然后比较2种训练模式下,每种edge的auc/f1(paper里那种方式),发现3种edge一起训练的效果然而不如每种edge单独训练。请问这样的话,多种edge type意义何在?

Take only 300 person-item data for experiments, set the meta path to person-item-person, the number of walks and length are the default values, but this will give an error line...

作者: 您好!非常感谢你的开源工作,我在我的数据集上实现了这个模型,不过有个小疑问想和你探讨一下,我在计算两两item之间的相似度的时候,发现相似度几乎是0.99以上,这种现象正常吗?我之前好像也看到有人提出类似的问题。以下是我取其中某一部分数据top10 ![image](https://user-images.githubusercontent.com/35071040/92367442-fafb9900-f128-11ea-8421-7ffb64366b57.png)

最近阅读了您的文章《Representation Learning for Atributed Multiplex Heterogeneous Network》,对其中的GATNE-I模型十分的感兴趣。想要用自己的数据集来进行实验,该模型是可以学习新的节点的,那么请问该模型在数据集构造时也需要保证valid.txt和test.txt中的节点必须下train.txt中存在嘛。

Hello! I would like to ask how to construct the training set, verification set and test set in the article code. Is there a specific code or idea?

Hello, Actually, I'm a social scientist. Regarding this, I'm trying to deploy your work to analyse and forecast social interaction. Therefore, the number of nodes is tremendously large in my...