What does xmp.MpSerialExecutor() really do?

xla
1

Good day, I’m kinda new to using pytorch xla. In the colab tutorial provided serial executor was used. I was just wondering in what other scenario can it be used

 # Using the serial executor avoids multiple processes to
  # download the same data.
  train_dataset, test_dataset = SERIAL_EXEC.run(get_dataset)

  train_sampler = torch.utils.data.distributed.DistributedSampler(
    train_dataset,
    num_replicas=xm.xrt_world_size(),
    rank=xm.get_ordinal(),
    shuffle=True)
  train_loader = torch.utils.data.DataLoader(
      train_dataset,
      batch_size=FLAGS['batch_size'],
      sampler=train_sampler,
      num_workers=FLAGS['num_workers'],
      drop_last=True)
  test_loader = torch.utils.data.DataLoader(
      test_dataset,
      batch_size=FLAGS['batch_size'],
      shuffle=False,
      num_workers=FLAGS['num_workers'],
      drop_last=True)