Documentation

Setting drop_remainder for Static Batching

Navigate to the data preprocessing file official/vision/image_classification/resnet/imagenet_preprocessing.py based on the training script logic.

Set drop_remainder to True in the data read function input_fn to complete the porting.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
def input_fn(is_training,
             data_dir,
             batch_size,
             dtype=tf.float32,
             datasets_num_private_threads=None,
             parse_record_fn=parse_record,
             input_context=None,
             drop_remainder=False,
             tf_data_experimental_slack=False,
             training_dataset_cache=False,
             filenames=None):
......

Returns:
  A dataset that can be used for iteration.
"""
drop_remainder=True

if filenames is None:
  filenames = get_filenames(is_training, data_dir)
dataset = tf.data.Dataset.from_tensor_slices(filenames)