paddle 2.3.x 报错 ValueError: `type` to initialized an Operator can not be None.

Traceback (most recent call last):

  File ~\Desktop\YT_work\sc\order\退费w2v.py:62 in <module>
    vecs, = exe.run(

  File ~\anaconda3\lib\site-packages\paddle\fluid\executor.py:1299 in run
    six.reraise(*sys.exc_info())

  File ~\anaconda3\lib\site-packages\six.py:719 in reraise
    raise value

  File ~\anaconda3\lib\site-packages\paddle\fluid\executor.py:1285 in run
    res = self._run_impl(

  File ~\anaconda3\lib\site-packages\paddle\fluid\executor.py:1431 in _run_impl
    program = self._add_feed_fetch_ops(

  File ~\anaconda3\lib\site-packages\paddle\fluid\executor.py:767 in _add_feed_fetch_ops
    tmp_program = program.clone()

  File ~\anaconda3\lib\site-packages\paddle\fluid\framework.py:5443 in clone
    p._sync_with_cpp()

  File ~\anaconda3\lib\site-packages\paddle\fluid\framework.py:5983 in _sync_with_cpp
    block._sync_with_cpp()

  File ~\anaconda3\lib\site-packages\paddle\fluid\framework.py:3771 in _sync_with_cpp
    op = Operator(self, op_desc)

  File ~\anaconda3\lib\site-packages\paddle\fluid\framework.py:2594 in __init__
    raise ValueError(

ValueError: `type` to initialized an Operator can not be None.

解决方法:

开头加入该代码（use_gpu=True时，我的电脑因没有gpu会报错，但这样之后上面的错误却不会出现了）

import paddlehub as hub
#  该处运行一遍后注销
try:
    lac = hub.Module(name="lac")
    results = lac.lexical_analysis(data={"text" : ["国王", "王后"]}, use_gpu=True, batch_size=2)
except:pass

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~