文本纠错AssertionError

[wannian1997]

checker.py文件中50行assert len(mask_id) == 1出现AssertionError请问如何解决

[wannian1997]

我将这句改为了assert len(mask_id) >= 1，但是这样输出的结果为类似{(8, ''): [('[', 0.8247228860855103), (']', 0.08479063957929611), ('被', 0.006882567889988422), ('/', 0.0036284865345805883), ('『', 0.0036059825215488672)]}，缺少了原本的错误的字。

[SeanLee97]

将你的输入文本和测试代码也发一下？不然不好追踪原因。

[wannian1997]

'import xmnlp
modelPath = r"E:\deepLearning\xmnlp-onnx-models"
xmnlp.set_model(modelPath)
text = "不能适应体育专业选拔人材的要求"
print(xmnlp.checker(text))'

Lazy load checker...
Traceback (most recent call last):
File "E:/deepLearning/Bert-Chinese-Text-Classification-Pytorch/toolTest.py", line 6, in
print(xmnlp.checker(text))
File "D:\Programs\Anaconda\envs\py36\lib\site-packages\xmnlp\checker_init_.py", line 40, in spellcheck
load_checker()
File "D:\Programs\Anaconda\envs\py36\lib\site-packages\xmnlp\checker_init_.py", line 29, in load_checker
os.path.join(config.MODEL_DIR, 'checker'))
File "D:\Programs\Anaconda\envs\py36\lib\site-packages\xmnlp\checker\checker.py", line 50, in init
assert len(mask_id) == 1
AssertionError

进程已结束,退出代码1

[jiluojiluo]

import xmnlp
text = "不能适应体育专业选拔人材的要求"
File "", line 1
text = "不能适应体育专业选拔人材的要求"
IndentationError: unexpected indent
text = "不能适应体育专业选拔人材的要求"
print(xmnlp.checker(text))
(Lazy Load) Loading model...
start to build dictionary...
done !
Traceback (most recent call last):
File "", line 1, in
File "C:\Users\l00660157\Anaconda3\envs\labelme\Lib\site-packages\xmnlp_init_.py", line 135, in checker
return checker.check(text, level=level)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\l00660157\Anaconda3\envs\labelme\Lib\site-packages\xmnlp\checker_init.py", line 48, in check
return model.best_match(doc)
^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\l00660157\Anaconda3\envs\labelme\Lib\site-packages\xmnlp\checker\checker.py", line 110, in best_match
return self.word_checker(word)[0]
~~~~~~~~~~~~~~~~~~~~~~~^^^
TypeError: 'NoneType' object is not subscriptable

[jiluojiluo]

我的报这个错误

TypeError: 'NoneType' object is not subscriptable