After the exercise of building convolutional, RNN, sentence level attention RNN, finally I have come to implement Hierarchical Attention Networks for Document Classification. I'm very thankful to Keras, which make building this project painless. The
本文记录笔者在观看Coursera上国立经济大学HLE的课程 How to win a data science competetion中的收获,和大家分享.课程的这门课的讲授人是Kaggle的大牛,内容详实,需要一定的python和机器学习基础.笔者只记录了一部分,许多细节都被省略了,详细内容各位看官还请移步课程链接 1. Week1 Feature Preprocessing 特征预处理 数值型 Numerical Tree-Based 模型不依赖于归一化(scaling),Non-tree