DecodingResult

class opennmt.utils.DecodingResult(ids, lengths, log_probs, attention, state)[source]

Final decoding result.

Parameters

ids – The predicted ids of shape $[B, H, T_{t}]$ .
lengths – The produced sequences length of shape $[B, H]$ .
log_probs – The cumulated log probabilities of shape $[B, H]$ .
attention – The attention history of shape $[B, H, T_{t}, T_{s}]$ .
state – The final decoding state.

Inherits from: builtins.tuple