OpenNMT/OpenNMT-tf

Multi-decoder setup for multi-task learning

Open

#130 geöffnet am 22. Mai 2018

Auf GitHub ansehen
 (5 Kommentare) (0 Reaktionen) (0 zugewiesene Personen)Python (394 Forks)batch import
enhancementhelp wanted

Repository-Metriken

Stars
 (1.428 Stars)
PR-Merge-Metriken
 (Keine gemergten PRs in 30 T)

Beschreibung

Similarly to the ParallelEncoder, a ParallelDecoder setup could allow multi-task learning. This should not be too hard to implement but we need to take care of some details:

  • support separate values for the decoding parameters (beam_width, length_penalty, etc.),
  • parts of SequenceToSequence assume a single output head (e.g. loss computation, reverse vocabulary lookup, exported outputs for model serving, etc. which should be moved in the decoder itself)

Contributor Guide