Python 使用数据扩充丢失实例

我试图从TF数据集中扩充数据。我使用了TF文档（）中的代码。我的问题是，在扩展之后，我剩下的实例比开始时少。我的目标是向原始数据中添加增强数据。这是我的密码：

(train_ds, val_ds, test_ds), metadata = tfds.load(
    'rock_paper_scissors',
    split=['train[:80%]', 'train[80%:90%]', 'train[90%:]'],
    shuffle_files = False,
    with_info=True,
    as_supervised=True,
)

# Check the number of examples in each dataset to ensure the data split/loading worked.
train_size = tf.data.experimental.cardinality(train_ds)
valid_size = tf.data.experimental.cardinality(val_ds)
test_size = tf.data.experimental.cardinality(test_ds)
print ('train: %d, valid: %d, test: %d' % (train_size, valid_size, test_size))
# should be train: 2016, valid: 504, test: 372

train: 2016, valid: 252, test: 252

IMG_SIZE = 299

resize_and_rescale = tf.keras.Sequential([
  tf.keras.layers.experimental.preprocessing.Resizing(IMG_SIZE, IMG_SIZE),
  tf.keras.layers.experimental.preprocessing.Rescaling(1./255)
])

data_augmentation = tf.keras.Sequential([
  tf.keras.layers.experimental.preprocessing.RandomFlip("horizontal_and_vertical"),
  tf.keras.layers.experimental.preprocessing.RandomRotation(0.2),
])

batch_size = 32
AUTOTUNE = tf.data.experimental.AUTOTUNE

def prepare(ds, shuffle=False, augment=False):
  # Resize and rescale all datasets
  ds = ds.map(lambda x, y: (resize_and_rescale(x), y), 
              num_parallel_calls=AUTOTUNE)

  if shuffle:
    ds = ds.shuffle(1000)

  # Batch all datasets
  ds = ds.batch(batch_size)

  # Use data augmentation only on the training set
  if augment:
    ds = ds.map(lambda x, y: (data_augmentation(x, training=True), y), 
                num_parallel_calls=AUTOTUNE)

  # Use buffered prefecting on all datasets
  return ds.prefetch(buffer_size=AUTOTUNE)

train_ds = prepare(train_ds, shuffle=True, augment=True)
val_ds = prepare(val_ds)
test_ds = prepare(test_ds)

# Check the number of examples in each dataset to ensure the data split/loading worked.
train_size = tf.data.experimental.cardinality(train_ds)
valid_size = tf.data.experimental.cardinality(val_ds)
test_size = tf.data.experimental.cardinality(test_ds)
print ('train: %d, valid: %d, test: %d' % (train_size, valid_size, test_size))

train: 63, valid: 8, test: 8