c9b1d78f806892193efdebee2789a47fd148b984,examples/imagenet_preprocessing.py,,input_fn,#Any#Any#Any#Any#Any#Any#Any#Any#Any#Any#,477

Before Change

  // This number is low enough to not cause too much contention on small systems
  // but high enough to provide the benefits of parallelization. You may want
  // to increase this number if you have a large number of CPU cores.
  dataset = dataset.apply(tf.data.experimental.parallel_interleave(
      tf.data.TFRecordDataset, cycle_length=48))

  return process_record_dataset(
      dataset=dataset,
      is_training=is_training,

After Change

  // This number is low enough to not cause too much contention on small systems
  // but high enough to provide the benefits of parallelization. You may want
  // to increase this number if you have a large number of CPU cores.
  dataset = dataset.interleave(
      tf.data.TFRecordDataset, cycle_length=48,
      num_parallel_calls=tf.data.experimental.AUTOTUNE)

  return process_record_dataset(
      dataset=dataset,
      is_training=is_training,

Instances