Attention机制在NLP上最早是被用于seq2seq的翻译类任务中，如Neural Machine Translation by Jointly Learning to Align and TranslateNeura
阅读背景：
用于文本分类的RNN-Attention网络
来源：互联网
分享到：
