多标签版本KNN，TypeError: __init__() takes 1 positional argument but 2 were given

在使用MLKNN时，传递x_train和y_train时报错TypeError: init() takes 1 positional argument but 2 were given

我的代码

classifier = MLkNN()
classifier.fit(features,labels)
pred = classifier.predict(x_test)
print(accuracy_score(y_test,pred))

报错信息
Traceback (most recent call last):
File "E:/python项目/LEAM-master 2/test.py", line 26, in
classifier.fit(features,labels)
File "C:\ProgramData\Anaconda3\envs\LEAM-master 2\lib\site-packages\skmultilearn\adapt\mlknn.py", line 218, in fit
self._cond_prob_true, self._cond_prob_false = self._compute_cond(X, self._label_cache)
File "C:\ProgramData\Anaconda3\envs\LEAM-master 2\lib\site-packages\skmultilearn\adapt\mlknn.py", line 165, in compute_cond
self.knn = NearestNeighbors(self.k).fit(X)
TypeError: init() takes 1 positional argument but 2 were given

这个是参数对不上问题，你看看你classifier.fit(features,labels)的fit源码有多小个参数，导致参数不同，应该是新旧版本的变更问题

同一个错误，我本来其实是可以正常运行的，但是有一次我升级了scikit-learn的版本，然后就出现博主这个错误，后面我把版本降回来，就不报错了，我电脑这里不报错时，scikit-learn的版本是0.23.2

楼主怎么解决的呢？label_binarize() takes 1 positional argument but 2 were given