search

limaosen0 / Variational-Graph-Auto-Encoders

This is the implementation of paper 'Variational Graph Auto-Encoder' in NIPS Workshop on Bayesian Deep Learning, 2016.
74 stars 23 forks source link

about the 'n_embedding' #1

Open george-kris opened 5 years ago

george-kris commented 5 years ago

how to initialize the n_embedding from arg? thank you !

johnqoe commented 5 years ago

how to initialize the n_embedding from arg? thank you !

Have you solved this problem? I also encountered this problem.

george-kris commented 5 years ago

no. but i found the original code provided by the GAE's author. https://github.com/tkipf/gae if you want to use the code provided by limaosen0, i guess you need train the graph first, get the embedding an use it.

------------------ 原始邮件 ------------------ 发件人: "johnqoe"notifications@github.com; 发送时间: 2019年3月16日(星期六) 下午5:18 收件人: "limaosen0/Variational-Graph-Auto-Encoders"Variational-Graph-Auto-Encoders@noreply.github.com; 抄送: "木子"1228195997@qq.com; "Author"author@noreply.github.com; 主题: Re: [limaosen0/Variational-Graph-Auto-Encoders] about the'n_embedding' (#1)

how to initialize the n_embedding from arg? thank you !

Have you solved this problem? I also encountered this problem.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.

johnqoe commented 5 years ago

no. but i found the original code provided by the GAE's author. https://github.com/tkipf/gae if you want to use the code provided by limaosen0, i guess you need train the graph first, get the embedding an use it. … ------------------ 原始邮件 ------------------ 发件人: "johnqoe"notifications@github.com; 发送时间: 2019年3月16日(星期六) 下午5:18 收件人: "limaosen0/Variational-Graph-Auto-Encoders"Variational-Graph-Auto-Encoders@noreply.github.com; 抄送: "木子"1228195997@qq.com; "Author"author@noreply.github.com; 主题: Re: [limaosen0/Variational-Graph-Auto-Encoders] about the'n_embedding' (#1) how to initialize the n_embedding from arg? thank you ! Have you solved this problem? I also encountered this problem. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.

Great，thank you very much.

Billy1900 commented 4 years ago

n_embedding stands for the number of the implicit units, it should be a constant number, usually, we set 128.