espnet.utils.training.batchfy.batchfy_by_seq

source

espnet.utils.training.batchfy.batchfy_by_seq(sorted_data, batch_size, max_length_in, max_length_out, min_batch_size=1, shortest_first=False, ikey='input', iaxis=0, okey='output', oaxis=0)

Make batch set from json dictionary.

• Parameters:
  • sorted_data (Dict *[*str , Dict *[*str , Any ] ]) – dictionary loaded from data.json
  • batch_size (int) – batch size
  • max_length_in (int) – maximum length of input to decide adaptive batch size
  • max_length_out (int) – maximum length of output to decide adaptive batch size
  • min_batch_size (int) – mininum batch size (for multi-gpu)
  • shortest_first (bool) – Sort from batch with shortest samples to longest if true, otherwise reverse
  • ikey (str) – key to access input (for ASR ikey=”input”, for TTS, MT ikey=”output”.)
  • iaxis (int) – dimension to access input (for ASR, TTS iaxis=0, for MT iaxis=”1”.)
  • okey (str) – key to access output (for ASR, MT okey=”output”. for TTS okey=”input”.)
  • oaxis (int) – dimension to access output (for ASR, TTS, MT oaxis=0, reserved for future research, -1 means all axis.)
• Returns: List[List[Tuple[str, dict]]] list of batches