medium_training_dataset.py

def build_training_dataset():

    pairs_tensor = tf.convert_to_tensor(training_pairs)
    labels_tensor = tf.convert_to_tensor(training_pairs_labels)

    result = tf.data.Dataset.from_tensor_slices((pairs_tensor, labels_tensor))

    result = result.map(lambda pair, label: (load_images(pair), label))
    result = result.shuffle(100, reshuffle_each_iteration=True)
    result = result.repeat()
    result = result.batch(TRAINING_BATCH_SIZE)
    result = result.map(lambda pair, y: ((data_augmentation(pair[0], training=True),data_augmentation(pair[1], training=True)), y), 
                num_parallel_calls=tf.data.AUTOTUNE)

    result = result.prefetch(tf.data.AUTOTUNE)

    return result

train_ds = build_training_dataset()