This setting is commonly used in the encoder-decoder sequence-to-sequence model, where the encoder final state is used as the initial state ...
確定! 回上一頁