Method decode

examples/server/server.py:13055–13085

(
            cls,
            *,
            request_id: str,
            seq_id: int,
            start_pos: int,
            llama_start_pos: int,
            tokens: List[int],
            identity_tokens: List[int],
            output_indices: List[Optional[int]],
            output_positions: List[int],
            position_increments: List[int],
            completion_index: int,
            pending_count: int,
        )

        @classmethod
        def decode(
            cls,
            *,
            request_id: str,
            seq_id: int,
            start_pos: int,
            llama_start_pos: int,
            tokens: List[int],
            identity_tokens: List[int],
            output_indices: List[Optional[int]],
            output_positions: List[int],
            position_increments: List[int],
            completion_index: int,
            pending_count: int,
        ) -> "CompletionScheduler.BatchItem":
            return cls(
                kind="decode",
                request_id=request_id,
                seq_id=seq_id,
                start_pos=start_pos,
                llama_start_pos=llama_start_pos,
                tokens=tokens,
                identity_tokens=identity_tokens,
                output_indices=output_indices,
                output_positions=output_positions,
                position_increments=position_increments,
                decode_state=cls.Decode(
                    completion_index=completion_index,
                    pending_count=pending_count,
                ),
            )

    def __init__(
        self,