【TensorFlow】如何判断当前进行到第几个epoch了？

def read_and_decode_TFRecordDataset_deblur(tfrecords_path, batch_size, epoch_num):
    dataset = tf.data.TFRecordDataset(tfrecords_path)
    dataset = dataset.map(parser_deblur).shuffle(buffer_size=100*batch_size)
    epoch = tf.data.Dataset.range(epoch_num)
    dataset = epoch.flat_map(lambda i: tf.data.Dataset.zip(
        (dataset, tf.data.Dataset.from_tensors(i).repeat())))
    dataset = dataset.batch(batch_size)
    iterator = dataset.make_one_shot_iterator()
    (face_blur_batch, face_gt_batch), epochNow = iterator.get_next()
    return face_blur_batch, face_gt_batch, epochNow

其中有2点比较重要：

1. 一定要在指定epoch之前shuffle()

REFERENCE: How to set epoch counter when using TFRecordDataset?

2. shuffle的buffer_size如何设置

REFERENCE: Meaning of buffer_size in Dataset.map , Dataset.prefetch and Dataset.shuffle

REFERENCE: tf.data.Dataset.shuffle(buffer_size)中buffer_size的理解