New answers tagged text-generation
3
After a Googling around, I think this tutorial may suit your needs.
However, it seems you have a misconception about the Transformer decoder: in training mode there is no iteration at all. While LSTM-based decoders are autoregressive by nature, Transformers are not. Instead, all predictions are generated at once based on the real target tokens (i.e. teacher ...
Top 50 recent answers are included
