XufengXufengXufeng / try_gcn

try different opts on word context graph with GCN and GAT to obtain word embeddings.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

try_gcn

compare node2vec, gcn, gat on word context graph for word embeddings.

尝试使用GCN做词向量。没有node2vec效果好,但是比随机的好。然后又做了GAT,比GCN拟合慢,而且容易进入局部最优。

嵌入方法:

  1. 完全随机的参数。(GCN)
  2. 监督预测结巴词性。(GCN,GAT)
  3. 监督预测(skip-gram)词上下文。(GCN,GAT)

About

try different opts on word context graph with GCN and GAT to obtain word embeddings.


Languages

Language:Jupyter Notebook 100.0%