3.1 Input; 3.2 Encoder–decoder architecture; 3.3 Scaled dot-product attention. 3.3.1 Multi-head attention; 3.3.2 Masked attention. 3.4 Encoder.
確定! 回上一頁