图卷积网络（GCN）python实现
时间:2022-07-23
本文章向大家介绍图卷积网络（GCN）python实现，主要内容包括其使用实例、应用技巧、基本知识点总结和需要注意事项，具有一定的参考价值，需要的朋友可以参考一下。
数据集为cora数据集，cora数据集由机器学习论文组成，共以下7类：
- 基于案例
- 遗传算法
- 神经网络
- 概率方法
- 强化学习
- 规则学习
- 理论
由cora.content和cora.cities文件构成。共2708个样本，每个样本的特征维度是1433。
下载地址：https://linqs.soe.ucsc.edu/data
cora.content：
每一行由论文id+特征向量+标签构成。
/*
* 提示：该行代码过长，系统自动注释不进行高亮。一键复制会移除系统注释
* 31336 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 Neural_Networks
*/
/*
* 提示：该行代码过长，系统自动注释不进行高亮。一键复制会移除系统注释
* 1061127 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 Rule_Learning
*/
/*
* 提示：该行代码过长，系统自动注释不进行高亮。一键复制会移除系统注释
* 1106406 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 Reinforcement_Learning
*/
/*
* 提示：该行代码过长，系统自动注释不进行高亮。一键复制会移除系统注释
* 13195 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Reinforcement_Learning
*/
......cora.cities：
引用关系：被引论文编号以及引论文编号
35 1033
35 103482
35 103515
35 1050679
35 1103960
35 1103985
35 1109199
35 1112911
......读取数据集：
import numpy as np
import scipy.sparse as sp
import torch
from sklearn.preprocessing import LabelBinarizer
def normalize_adj(adjacency):
adjacency += sp.eye(adjacency.shape[0])
degree = np.array(adjacency.sum(1))
d_hat = sp.diags(np.power(degree, -0.5).flatten())
return d_hat.dot(adjacency).dot(d_hat).tocoo()
def normalize_features(features):
return features / features.sum(1)
def load_data(path="/content/drive/My Drive/nlpdata/cora/", dataset="cora"):
"""Load citation network dataset (cora only for now)"""
print('Loading {} dataset...'.format(dataset))
idx_features_labels = np.genfromtxt("{}{}.content".format(path,dataset), dtype=np.dtype(str))
features = sp.csr_matrix(idx_features_labels[:, 1:-1], dtype=np.float32)
encode_onehot = LabelBinarizer()
labels = encode_onehot.fit_transform(idx_features_labels[:, -1])
# build graph
idx = np.array(idx_features_labels[:, 0], dtype=np.int32)
idx_map = {j: i for i, j in enumerate(idx)}
edges_unordered = np.genfromtxt("{}{}.cites".format(path, dataset), dtype=np.int32)
edges = np.array(list(map(idx_map.get, edges_unordered.flatten())), dtype=np.int32).reshape(edges_unordered.shape)
adj = sp.coo_matrix((np.ones(edges.shape[0]), (edges[:, 0], edges[:, 1])), shape=(labels.shape[0], labels.shape[0]), dtype=np.float32)
features = normalize_features(features)
adj = normalize_adj(adj)
idx_train = range(140)
idx_val = range(200, 500)
idx_test = range(500, 1500)
features = torch.FloatTensor(np.array(features))
labels = torch.LongTensor(np.where(labels)[1])
num_nodes = features.shape[0]
train_mask = np.zeros(num_nodes, dtype=np.bool)
val_mask = np.zeros(num_nodes, dtype=np.bool)
test_mask = np.zeros(num_nodes, dtype=np.bool)
train_mask[idx_train] = True
val_mask[idx_val] = True
test_mask[idx_test] = True
return adj, features, labels, train_mask, val_mask, test_mask构建网络：
import numpy as np
import scipy.sparse as sp
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.nn.init as init
import torch.optim as optim
import matplotlib.pyplot as plt
from load_cora import *
import sys
sys.path.append("/content/drive/My Drive/nlpdata/cora/")
class GraphConvolution(nn.Module):
def __init__(self, input_dim, output_dim, use_bias=True):
"""图卷积：L*X*theta
Args:
----------
input_dim: int
节点输入特征的维度
output_dim: int
输出特征维度
use_bias : bool, optional
是否使用偏置
"""
super(GraphConvolution, self).__init__()
self.input_dim = input_dim
self.output_dim = output_dim
self.use_bias = use_bias
self.weight = nn.Parameter(torch.Tensor(input_dim, output_dim))
if self.use_bias:
self.bias = nn.Parameter(torch.Tensor(output_dim))
else:
self.register_parameter('bias', None)
self.reset_parameters()
def reset_parameters(self):
init.kaiming_uniform_(self.weight)
if self.use_bias:
init.zeros_(self.bias)
def forward(self, adjacency, input_feature):
"""邻接矩阵是稀疏矩阵，因此在计算时使用稀疏矩阵乘法
Args:
-------
adjacency: torch.sparse.FloatTensor
邻接矩阵
input_feature: torch.Tensor
输入特征
"""
device = "cuda" if torch.cuda.is_available() else "cpu"
support = torch.mm(input_feature, self.weight.to(device))
output = torch.sparse.mm(adjacency, support)
if self.use_bias:
output += self.bias.to(device)
return output
def __repr__(self):
return self.__class__.__name__ + ' (' + str(self.in_features) + ' -> ' + str(self.out_features) + ')'
# ## 模型定义
class GcnNet(nn.Module):
"""
定义一个包含两层GraphConvolution的模型
"""
def __init__(self, input_dim=1433):
super(GcnNet, self).__init__()
self.gcn1 = GraphConvolution(input_dim, 16)
self.gcn2 = GraphConvolution(16, 7)
def forward(self, adjacency, feature):
h = F.relu(self.gcn1(adjacency, feature))
logits = self.gcn2(adjacency, h)
return logits进行训练和测试：
# ## 模型训练
# 超参数定义
learning_rate = 0.1
weight_decay = 5e-4
epochs = 200
# 模型定义：Model, Loss, Optimizer
device = "cuda" if torch.cuda.is_available() else "cpu"
model = GcnNet().to(device)
criterion = nn.CrossEntropyLoss().to(device)
optimizer = optim.Adam(model.parameters(), lr=learning_rate, weight_decay=weight_decay)
adjacency, features, labels, train_mask, val_mask, test_mask= load_data()
tensor_x = features.to(device)
tensor_y = labels.to(device)
tensor_train_mask = torch.from_numpy(train_mask).to(device)
tensor_val_mask = torch.from_numpy(val_mask).to(device)
tensor_test_mask = torch.from_numpy(test_mask).to(device)
indices = torch.from_numpy(np.asarray([adjacency.row, adjacency.col]).astype('int64')).long()
values = torch.from_numpy(adjacency.data.astype(np.float32))
tensor_adjacency = torch.sparse.FloatTensor(indices, values, (2708, 2708)).to(device)
# 训练主体函数
def train():
loss_history = []
val_acc_history = []
model.train()
train_y = tensor_y[tensor_train_mask]
for epoch in range(epochs):
logits = model(tensor_adjacency, tensor_x) # 前向传播
train_mask_logits = logits[tensor_train_mask] # 只选择训练节点进行监督
loss = criterion(train_mask_logits, train_y) # 计算损失值
optimizer.zero_grad()
loss.backward() # 反向传播计算参数的梯度
optimizer.step() # 使用优化方法进行梯度更新
train_acc, _, _ = test(tensor_train_mask) # 计算当前模型训练集上的准确率
val_acc, _, _ = test(tensor_val_mask) # 计算当前模型在验证集上的准确率
# 记录训练过程中损失值和准确率的变化，用于画图
loss_history.append(loss.item())
val_acc_history.append(val_acc.item())
print("Epoch {:03d}: Loss {:.4f}, TrainAcc {:.4}, ValAcc {:.4f}".format(
epoch, loss.item(), train_acc.item(), val_acc.item()))
return loss_history, val_acc_history
# 测试函数
def test(mask):
model.eval()
with torch.no_grad():
logits = model(tensor_adjacency, tensor_x)
test_mask_logits = logits[mask]
predict_y = test_mask_logits.max(1)[1]
accuarcy = torch.eq(predict_y, tensor_y[mask]).float().mean()
return accuarcy, test_mask_logits.cpu().numpy(), tensor_y[mask].cpu().numpy()
if __name__ == "__main__":
train()
test_accuracy, _, _ = test(tensor_test_mask)
print("测试准确率是：{:.4f}".format(test_accuracy))结果：
Loading cora dataset...
Epoch 000: Loss 1.9681, TrainAcc 0.3286, ValAcc 0.3467
Epoch 001: Loss 1.7307, TrainAcc 0.4786, ValAcc 0.4033
Epoch 002: Loss 1.5521, TrainAcc 0.5214, ValAcc 0.4033
Epoch 003: Loss 1.3685, TrainAcc 0.6143, ValAcc 0.5100
Epoch 004: Loss 1.1594, TrainAcc 0.75, ValAcc 0.5767
Epoch 005: Loss 0.9785, TrainAcc 0.7857, ValAcc 0.5900
Epoch 006: Loss 0.8226, TrainAcc 0.8286, ValAcc 0.5867
Epoch 007: Loss 0.6849, TrainAcc 0.8929, ValAcc 0.6200
Epoch 008: Loss 0.5448, TrainAcc 0.9429, ValAcc 0.6433
Epoch 009: Loss 0.4152, TrainAcc 0.9429, ValAcc 0.6667
Epoch 010: Loss 0.3221, TrainAcc 0.9857, ValAcc 0.6767
Epoch 011: Loss 0.2547, TrainAcc 1.0, ValAcc 0.7033
Epoch 012: Loss 0.1979, TrainAcc 1.0, ValAcc 0.7167
Epoch 013: Loss 0.1536, TrainAcc 1.0, ValAcc 0.7000
Epoch 014: Loss 0.1276, TrainAcc 1.0, ValAcc 0.6700
Epoch 015: Loss 0.1122, TrainAcc 1.0, ValAcc 0.6867
Epoch 016: Loss 0.0979, TrainAcc 1.0, ValAcc 0.6800
Epoch 017: Loss 0.0876, TrainAcc 1.0, ValAcc 0.6700
Epoch 018: Loss 0.0821, TrainAcc 1.0, ValAcc 0.6667
Epoch 019: Loss 0.0799, TrainAcc 1.0, ValAcc 0.6800
Epoch 020: Loss 0.0804, TrainAcc 1.0, ValAcc 0.6933
Epoch 021: Loss 0.0852, TrainAcc 1.0, ValAcc 0.6833
Epoch 022: Loss 0.0904, TrainAcc 1.0, ValAcc 0.6700
Epoch 023: Loss 0.0914, TrainAcc 1.0, ValAcc 0.6700
Epoch 024: Loss 0.0926, TrainAcc 1.0, ValAcc 0.6400
Epoch 025: Loss 0.0953, TrainAcc 1.0, ValAcc 0.6300
Epoch 026: Loss 0.0931, TrainAcc 1.0, ValAcc 0.6467
Epoch 027: Loss 0.0880, TrainAcc 1.0, ValAcc 0.6600
Epoch 028: Loss 0.0851, TrainAcc 1.0, ValAcc 0.6567
Epoch 029: Loss 0.0814, TrainAcc 1.0, ValAcc 0.6600
Epoch 030: Loss 0.0756, TrainAcc 1.0, ValAcc 0.6433
Epoch 031: Loss 0.0709, TrainAcc 1.0, ValAcc 0.6567
Epoch 032: Loss 0.0682, TrainAcc 1.0, ValAcc 0.6467
Epoch 033: Loss 0.0656, TrainAcc 1.0, ValAcc 0.6700
Epoch 034: Loss 0.0628, TrainAcc 1.0, ValAcc 0.6633
Epoch 035: Loss 0.0618, TrainAcc 1.0, ValAcc 0.6833
Epoch 036: Loss 0.0617, TrainAcc 1.0, ValAcc 0.6700
Epoch 037: Loss 0.0615, TrainAcc 1.0, ValAcc 0.6667
Epoch 038: Loss 0.0615, TrainAcc 1.0, ValAcc 0.6600
Epoch 039: Loss 0.0616, TrainAcc 1.0, ValAcc 0.6733
Epoch 040: Loss 0.0613, TrainAcc 1.0, ValAcc 0.6800
Epoch 041: Loss 0.0610, TrainAcc 1.0, ValAcc 0.6867
Epoch 042: Loss 0.0603, TrainAcc 1.0, ValAcc 0.6800
Epoch 043: Loss 0.0598, TrainAcc 1.0, ValAcc 0.6667
Epoch 044: Loss 0.0592, TrainAcc 1.0, ValAcc 0.6833
Epoch 045: Loss 0.0584, TrainAcc 1.0, ValAcc 0.6833
Epoch 046: Loss 0.0575, TrainAcc 1.0, ValAcc 0.6967
Epoch 047: Loss 0.0570, TrainAcc 1.0, ValAcc 0.6900
Epoch 048: Loss 0.0565, TrainAcc 1.0, ValAcc 0.6933
Epoch 049: Loss 0.0560, TrainAcc 1.0, ValAcc 0.6867
Epoch 050: Loss 0.0559, TrainAcc 1.0, ValAcc 0.6900
Epoch 051: Loss 0.0557, TrainAcc 1.0, ValAcc 0.6900
Epoch 052: Loss 0.0555, TrainAcc 1.0, ValAcc 0.6967
Epoch 053: Loss 0.0554, TrainAcc 1.0, ValAcc 0.6867
Epoch 054: Loss 0.0552, TrainAcc 1.0, ValAcc 0.6867
Epoch 055: Loss 0.0550, TrainAcc 1.0, ValAcc 0.6933
Epoch 056: Loss 0.0549, TrainAcc 1.0, ValAcc 0.7000
Epoch 057: Loss 0.0548, TrainAcc 1.0, ValAcc 0.7000
Epoch 058: Loss 0.0547, TrainAcc 1.0, ValAcc 0.7067
Epoch 059: Loss 0.0546, TrainAcc 1.0, ValAcc 0.7000
Epoch 060: Loss 0.0545, TrainAcc 1.0, ValAcc 0.6967
Epoch 061: Loss 0.0545, TrainAcc 1.0, ValAcc 0.6967
Epoch 062: Loss 0.0544, TrainAcc 1.0, ValAcc 0.7067
Epoch 063: Loss 0.0544, TrainAcc 1.0, ValAcc 0.7067
Epoch 064: Loss 0.0543, TrainAcc 1.0, ValAcc 0.7033
Epoch 065: Loss 0.0542, TrainAcc 1.0, ValAcc 0.7000
Epoch 066: Loss 0.0542, TrainAcc 1.0, ValAcc 0.7000
Epoch 067: Loss 0.0542, TrainAcc 1.0, ValAcc 0.7033
Epoch 068: Loss 0.0542, TrainAcc 1.0, ValAcc 0.7067
Epoch 069: Loss 0.0542, TrainAcc 1.0, ValAcc 0.7033
Epoch 070: Loss 0.0543, TrainAcc 1.0, ValAcc 0.7033
Epoch 071: Loss 0.0543, TrainAcc 1.0, ValAcc 0.7000
Epoch 072: Loss 0.0543, TrainAcc 1.0, ValAcc 0.7033
Epoch 073: Loss 0.0544, TrainAcc 1.0, ValAcc 0.7067
Epoch 074: Loss 0.0544, TrainAcc 1.0, ValAcc 0.7067
Epoch 075: Loss 0.0545, TrainAcc 1.0, ValAcc 0.7133
Epoch 076: Loss 0.0545, TrainAcc 1.0, ValAcc 0.7100
Epoch 077: Loss 0.0546, TrainAcc 1.0, ValAcc 0.7133
Epoch 078: Loss 0.0546, TrainAcc 1.0, ValAcc 0.7067
Epoch 079: Loss 0.0547, TrainAcc 1.0, ValAcc 0.7133
Epoch 080: Loss 0.0547, TrainAcc 1.0, ValAcc 0.7033
Epoch 081: Loss 0.0548, TrainAcc 1.0, ValAcc 0.7100
Epoch 082: Loss 0.0549, TrainAcc 1.0, ValAcc 0.7067
Epoch 083: Loss 0.0549, TrainAcc 1.0, ValAcc 0.7133
Epoch 084: Loss 0.0549, TrainAcc 1.0, ValAcc 0.7067
Epoch 085: Loss 0.0550, TrainAcc 1.0, ValAcc 0.7100
Epoch 086: Loss 0.0550, TrainAcc 1.0, ValAcc 0.7033
Epoch 087: Loss 0.0551, TrainAcc 1.0, ValAcc 0.7133
Epoch 088: Loss 0.0551, TrainAcc 1.0, ValAcc 0.7067
Epoch 089: Loss 0.0552, TrainAcc 1.0, ValAcc 0.7100
Epoch 090: Loss 0.0553, TrainAcc 1.0, ValAcc 0.6967
Epoch 091: Loss 0.0553, TrainAcc 1.0, ValAcc 0.7067
Epoch 092: Loss 0.0554, TrainAcc 1.0, ValAcc 0.6900
Epoch 093: Loss 0.0556, TrainAcc 1.0, ValAcc 0.7100
Epoch 094: Loss 0.0557, TrainAcc 1.0, ValAcc 0.6833
Epoch 095: Loss 0.0561, TrainAcc 1.0, ValAcc 0.7033
Epoch 096: Loss 0.0558, TrainAcc 1.0, ValAcc 0.6833
Epoch 097: Loss 0.0557, TrainAcc 1.0, ValAcc 0.7100
Epoch 098: Loss 0.0547, TrainAcc 1.0, ValAcc 0.7133
Epoch 099: Loss 0.0546, TrainAcc 1.0, ValAcc 0.6900
Epoch 100: Loss 0.0555, TrainAcc 1.0, ValAcc 0.7033
Epoch 101: Loss 0.0561, TrainAcc 1.0, ValAcc 0.6700
Epoch 102: Loss 0.0579, TrainAcc 1.0, ValAcc 0.6967
Epoch 103: Loss 0.0577, TrainAcc 1.0, ValAcc 0.6633
Epoch 104: Loss 0.0600, TrainAcc 1.0, ValAcc 0.7000
Epoch 105: Loss 0.0550, TrainAcc 1.0, ValAcc 0.6967
Epoch 106: Loss 0.0540, TrainAcc 1.0, ValAcc 0.6767
Epoch 107: Loss 0.0555, TrainAcc 1.0, ValAcc 0.6967
Epoch 108: Loss 0.0528, TrainAcc 1.0, ValAcc 0.7000
Epoch 109: Loss 0.0571, TrainAcc 1.0, ValAcc 0.6700
Epoch 110: Loss 0.0643, TrainAcc 1.0, ValAcc 0.6933
Epoch 111: Loss 0.0583, TrainAcc 1.0, ValAcc 0.6800
Epoch 112: Loss 0.0533, TrainAcc 1.0, ValAcc 0.6700
Epoch 113: Loss 0.0552, TrainAcc 1.0, ValAcc 0.7067
Epoch 114: Loss 0.0534, TrainAcc 1.0, ValAcc 0.6967
Epoch 115: Loss 0.0555, TrainAcc 1.0, ValAcc 0.6833
Epoch 116: Loss 0.0555, TrainAcc 1.0, ValAcc 0.6800
Epoch 117: Loss 0.0559, TrainAcc 1.0, ValAcc 0.6933
Epoch 118: Loss 0.0601, TrainAcc 1.0, ValAcc 0.6700
Epoch 119: Loss 0.0707, TrainAcc 1.0, ValAcc 0.6667
Epoch 120: Loss 0.0670, TrainAcc 1.0, ValAcc 0.6500
Epoch 121: Loss 0.0574, TrainAcc 1.0, ValAcc 0.6467
Epoch 122: Loss 0.0589, TrainAcc 1.0, ValAcc 0.7033
Epoch 123: Loss 0.0493, TrainAcc 1.0, ValAcc 0.6800
Epoch 124: Loss 0.0591, TrainAcc 1.0, ValAcc 0.6900
Epoch 125: Loss 0.0482, TrainAcc 1.0, ValAcc 0.6600
Epoch 126: Loss 0.0562, TrainAcc 1.0, ValAcc 0.6667
Epoch 127: Loss 0.0538, TrainAcc 1.0, ValAcc 0.6900
Epoch 128: Loss 0.0579, TrainAcc 1.0, ValAcc 0.6867
Epoch 129: Loss 0.0557, TrainAcc 1.0, ValAcc 0.6833
Epoch 130: Loss 0.0615, TrainAcc 1.0, ValAcc 0.6733
Epoch 131: Loss 0.0570, TrainAcc 1.0, ValAcc 0.6667
Epoch 132: Loss 0.0612, TrainAcc 1.0, ValAcc 0.6700
Epoch 133: Loss 0.0669, TrainAcc 1.0, ValAcc 0.6967
Epoch 134: Loss 0.0544, TrainAcc 1.0, ValAcc 0.6767
Epoch 135: Loss 0.0605, TrainAcc 1.0, ValAcc 0.6567
Epoch 136: Loss 0.0546, TrainAcc 1.0, ValAcc 0.6567
Epoch 137: Loss 0.0586, TrainAcc 1.0, ValAcc 0.7033
Epoch 138: Loss 0.0501, TrainAcc 1.0, ValAcc 0.6833
Epoch 139: Loss 0.0600, TrainAcc 1.0, ValAcc 0.7067
Epoch 140: Loss 0.0513, TrainAcc 1.0, ValAcc 0.6633
Epoch 141: Loss 0.0587, TrainAcc 1.0, ValAcc 0.6733
Epoch 142: Loss 0.0556, TrainAcc 1.0, ValAcc 0.6833
Epoch 143: Loss 0.0586, TrainAcc 1.0, ValAcc 0.6967
Epoch 144: Loss 0.0565, TrainAcc 1.0, ValAcc 0.6900
Epoch 145: Loss 0.0586, TrainAcc 1.0, ValAcc 0.6833
Epoch 146: Loss 0.0559, TrainAcc 1.0, ValAcc 0.6767
Epoch 147: Loss 0.0589, TrainAcc 1.0, ValAcc 0.6800
Epoch 148: Loss 0.0562, TrainAcc 1.0, ValAcc 0.6900
Epoch 149: Loss 0.0560, TrainAcc 1.0, ValAcc 0.6933
Epoch 150: Loss 0.0565, TrainAcc 1.0, ValAcc 0.6833
Epoch 151: Loss 0.0547, TrainAcc 1.0, ValAcc 0.6767
Epoch 152: Loss 0.0559, TrainAcc 1.0, ValAcc 0.6967
Epoch 153: Loss 0.0549, TrainAcc 1.0, ValAcc 0.6933
Epoch 154: Loss 0.0567, TrainAcc 1.0, ValAcc 0.7000
Epoch 155: Loss 0.0556, TrainAcc 1.0, ValAcc 0.6867
Epoch 156: Loss 0.0568, TrainAcc 1.0, ValAcc 0.6967
Epoch 157: Loss 0.0558, TrainAcc 1.0, ValAcc 0.6967
Epoch 158: Loss 0.0568, TrainAcc 1.0, ValAcc 0.6867
Epoch 159: Loss 0.0560, TrainAcc 1.0, ValAcc 0.6867
Epoch 160: Loss 0.0563, TrainAcc 1.0, ValAcc 0.6933
Epoch 161: Loss 0.0562, TrainAcc 1.0, ValAcc 0.6967
Epoch 162: Loss 0.0559, TrainAcc 1.0, ValAcc 0.6833
Epoch 163: Loss 0.0556, TrainAcc 1.0, ValAcc 0.7000
Epoch 164: Loss 0.0554, TrainAcc 1.0, ValAcc 0.7033
Epoch 165: Loss 0.0556, TrainAcc 1.0, ValAcc 0.6967
Epoch 166: Loss 0.0553, TrainAcc 1.0, ValAcc 0.6900
Epoch 167: Loss 0.0555, TrainAcc 1.0, ValAcc 0.7033
Epoch 168: Loss 0.0554, TrainAcc 1.0, ValAcc 0.6967
Epoch 169: Loss 0.0560, TrainAcc 1.0, ValAcc 0.6833
Epoch 170: Loss 0.0558, TrainAcc 1.0, ValAcc 0.6900
Epoch 171: Loss 0.0561, TrainAcc 1.0, ValAcc 0.7033
Epoch 172: Loss 0.0560, TrainAcc 1.0, ValAcc 0.6967
Epoch 173: Loss 0.0559, TrainAcc 1.0, ValAcc 0.6833
Epoch 174: Loss 0.0557, TrainAcc 1.0, ValAcc 0.6967
Epoch 175: Loss 0.0554, TrainAcc 1.0, ValAcc 0.7033
Epoch 176: Loss 0.0554, TrainAcc 1.0, ValAcc 0.7033
Epoch 177: Loss 0.0551, TrainAcc 1.0, ValAcc 0.6933
Epoch 178: Loss 0.0553, TrainAcc 1.0, ValAcc 0.6967
Epoch 179: Loss 0.0553, TrainAcc 1.0, ValAcc 0.7000
Epoch 180: Loss 0.0555, TrainAcc 1.0, ValAcc 0.6900
Epoch 181: Loss 0.0556, TrainAcc 1.0, ValAcc 0.7000
Epoch 182: Loss 0.0557, TrainAcc 1.0, ValAcc 0.7033
Epoch 183: Loss 0.0558, TrainAcc 1.0, ValAcc 0.6933
Epoch 184: Loss 0.0556, TrainAcc 1.0, ValAcc 0.6867
Epoch 185: Loss 0.0555, TrainAcc 1.0, ValAcc 0.7033
Epoch 186: Loss 0.0554, TrainAcc 1.0, ValAcc 0.7000
Epoch 187: Loss 0.0552, TrainAcc 1.0, ValAcc 0.6933
Epoch 188: Loss 0.0552, TrainAcc 1.0, ValAcc 0.6933
Epoch 189: Loss 0.0553, TrainAcc 1.0, ValAcc 0.7000
Epoch 190: Loss 0.0554, TrainAcc 1.0, ValAcc 0.7000
Epoch 191: Loss 0.0554, TrainAcc 1.0, ValAcc 0.6933
Epoch 192: Loss 0.0555, TrainAcc 1.0, ValAcc 0.7000
Epoch 193: Loss 0.0555, TrainAcc 1.0, ValAcc 0.7000
Epoch 194: Loss 0.0555, TrainAcc 1.0, ValAcc 0.6933
Epoch 195: Loss 0.0554, TrainAcc 1.0, ValAcc 0.6933
Epoch 196: Loss 0.0553, TrainAcc 1.0, ValAcc 0.7000
Epoch 197: Loss 0.0553, TrainAcc 1.0, ValAcc 0.7033
Epoch 198: Loss 0.0553, TrainAcc 1.0, ValAcc 0.6933
Epoch 199: Loss 0.0553, TrainAcc 1.0, ValAcc 0.7033
测试准确率是：0.6480最后是进行tsne降维可视化：
from sklearn.manifold import TSNE
test_accuracy, test_data, test_labels = test(tensor_test_mask)
tsne = TSNE(perplexity=30, n_components=2, init='pca', n_iter=5000) # TSNE降维，降到2
low_dim_embs = tsne.fit_transform(test_data)
plt.title('tsne result')
plt.scatter(low_dim_embs[:,0], low_dim_embs[:,1], marker='o', c=test_labels)
plt.savefig("tsne.png")参考：
https://blog.csdn.net/weixin_39373480/article/details/88742200
- sl 2.0 重要更新
- 云计算技术原理
- WCF技术剖析之五：利用ASP.NET兼容模式创建支持会话（Session）的WCF服务
- 进入AI时代，你准备好了吗？
- TiDB 在 G7 的实践和未来
- 投资钛值的你，知道钛链是什么吗？
- Is this a MS EnterLib DAAB BUG or not?
- Silverlight 2 has a Timer (DispatcherTimer)
- 难道调用ThreadPool.QueueUserWorkItem（）的时候，真是必须调用Thread.Sleep（N）吗？
- silverlight 《Hands-On-Labs》教程系列
- ASP.NET MVC5+EF6+EasyUI 后台管理系统（67）-MVC与ECharts
- 2018年机器学习和数据科学重要会议概览
- ASP.NET MVC5+EF6+EasyUI 后台管理系统（60）-系统总结
- WCF技术剖析之一：通过一个ASP.NET程序模拟WCF基础架构
- JavaScript 教程
- JavaScript 编辑工具
- JavaScript 与HTML
- JavaScript 与Java
- JavaScript 数据结构
- JavaScript 基本数据类型
- JavaScript 特殊数据类型
- JavaScript 运算符
- JavaScript typeof 运算符
- JavaScript 表达式
- JavaScript 类型转换
- JavaScript 基本语法
- JavaScript 注释
- Javascript 基本处理流程
- Javascript 选择结构
- Javascript if 语句
- Javascript if 语句的嵌套
- Javascript switch 语句
- Javascript 循环结构
- Javascript 循环结构实例
- Javascript 跳转语句
- Javascript 控制语句总结
- Javascript 函数介绍
- Javascript 函数的定义
- Javascript 函数调用
- Javascript 几种特殊的函数
- JavaScript 内置函数简介
- Javascript eval() 函数
- Javascript isFinite() 函数
- Javascript isNaN() 函数
- parseInt() 与 parseFloat()
- escape() 与 unescape()
- Javascript 字符串介绍
- Javascript length属性
- javascript 字符串函数
- Javascript 日期对象简介
- Javascript 日期对象用途
- Date 对象属性和方法
- Javascript 数组是什么
- Javascript 创建数组
- Javascript 数组赋值与取值
- Javascript 数组属性和方法
- 剑指 Offer 03. 数组中重复的数字
- 841. 钥匙和房间
- 一篇文章了解python常见内置异常报错
- 五分钟极速搭建kubernetes集群
- 没想到吧！关于Dubbo的『消费端线程池模型』官网也写错了。
- Go内存管理之代码的逃逸分析
- MySQL读锁的区别和应用场景分析
- Apache NIFI 讲解(读完立即入门)
- 了解NiFi最大线程池和处理器并发任务设置
- 深入理解 Apache NIFI Connection
- 【Go语言探险】线上奇怪日志问题的排查
- 探索 Apache NIFI 集群的高可用
- 面试被问：Spring Boot中Tomcat是如何启动的？
- Spring Boot 实现定时任务的动态增删启停等管理！
- Apache NIFI ExecuteScript组件脚本使用教程