Transformer 与BERT模型
Task 10
- Transformer的原理。
- BERT的原理。
- 利用预训练的BERT模型将句子转换为句向量,进行文本分类。
参考:
transformer github实现:https://github.com/Kyubyong/transformer
transformer pytorch分步实现:http://nlp.seas.harvard.edu/2018/04/03/attention.html
搞懂Transformer结构,看这篇PyTorch实现就够了:https://www.tinymind.cn/articles/3834
“变形金刚”为何强大:从模型到代码全面解析Google Tensor2Tensor系统:https://segmentfault.com/a/1190000015575985
bert理论:
bert系列1: https://medium.com/dissecting-bert/dissecting-be…
bert系列2: https://medium.com/dissecting-bert/dissecting-be…
bert系列3: https://medium.com/dissecting-bert/dissecting-be…
5 分钟入门 Google 最强NLP模型:BERT:https://www.jianshu.com/p/d110d0c13063
BERT – State of the Art Language Model for NLP: BERT – State of the Art Language Model for NLP htt