Datawhale《深度學習-NLP》Task10- Attention原理

參考： transformer github實現：GitHub - Kyubyong/transformer: A TensorFlow Implem... transformer pytorch分步實現：The Annotated Transformer 搞懂Transformer結構，看這篇PyTorch實現就夠了：搞懂Transformer結構，看這篇PyTorch實現就夠了！ - Tiny

>>阅读原文<<