我正在尝试从此处微调xView模型:https://github.com/DIUx-xView/baseline/releases
import tensorflow as tf
import os
checkpoint_meta_path = os.path.join("vanilla", "vanilla-ckpt.meta")
checkpoint_files = os.path.join("vanilla", "vanilla-ckpt")
graph = tf.Graph()
with graph.as_default():
with tf.Session() as sess:
saver = tf.train.import_meta_graph(checkpoint_meta_path)
但是我得到这个错误:
UnicodeDecodeError: 'utf-8' codec can't decode byte 0x80 in position 41: invalid start byte
我在做什么错?还是这与文件保存方式有关? GitHub上的此问题似乎表明应该存在一个检查点文件,并且可能导致该问题,但是xView并未提供此文件:https://github.com/tensorflow/tensorflow/issues/19573