Module seq2seq 

Available on crate feature std only.

Explicit aligned encoder-decoder examples and paired packed/padded collation. Paired encoder/decoder collation with explicit already-aligned teacher-forcing targets.

Structs

PackedSeq2SeqBatch

Native paired packed inputs with independent actual source/decoder boundaries.

PackedTokenBatch

Actual flat document inputs without labels or synthetic separator tokens.

PaddedSeq2SeqBatch

Native paired padded inputs and unchanged aligned decoder supervision.

PaddedTokenBatch

Actual padded token inputs without labels, for a model’s encoder or decoder.

Seq2SeqBatcher

Native paired collation without model-specific target construction or truncation.

Seq2SeqDataError

Invalid actual paired input/label geometry or explicit layout configuration.

Seq2SeqExample

One actual encoded source and its explicit decoder inputs and same-position labels.

Enums

Seq2SeqBatch

Explicit native paired device batch for padded or packed encoder-decoder models.

Seq2SeqBatchLayout

Paired document layout, with independently supplied source/decoder padding policies.