Skip to content

Latest commit

 

History

History
29 lines (19 loc) · 1.14 KB

readme.md

File metadata and controls

29 lines (19 loc) · 1.14 KB

station2vec

This repository provides an implementation of the VGAE (Variational Graph Autoencoder) for computing vector representations of stations.

model

I modified the Variational Graph Auto-Encoders architecture to optimize the following loss function;

$$loss = \mathbb{E} logp(X|Z) + \mathbb{E}[logp(A|Z)] - KL(q(Z|X, A)||p(Z))$$

where $KL[q(·)||p(·)]$ is the Kullback-Leibler divergence between $q(·)$ and $p(·)$. $A$ is an adjacency matrix. $X$ is a feature matrix. $Z$ is a latent vector.

利用データ

以下のデータを加工して作成(データは全て2023年8月30日取得)。

release

From the links below, you can search for similar or dissimilar stations.

https://atsushi-green.github.io/station2vec/