Tensorflow：读取后重写/保存graphdef

Question

我导入并编辑了一个冻结图，然后再次保存pb文件，如下所示。 由于解码错误，无法导入“ test2.pb”。 谁能解释一下并修复代码？

谢谢

filename = 'test.pb'
graph_def = tf.GraphDef()
with tf.gfile.GFile(filename, 'rb') as file:
    serialized_graph = file.read()
    graph_def.ParseFromString(serialized_graph)
    # Graph editing 
    ... 
    # end of graph editig 
    with tf.Session() as sess: 
        tf.train.write_graph(graph_def, "./", name='test2.pb')