.fit : AttributeError in python3，使用imblearn.ensemble和BalancedRandomForestClassifier

Question

代码：

from imblearn.ensemble import BalancedRandomForestClassifier

bal_forest = BalancedRandomForestClassifier(n_estimators=100, random_state=1)
bal_forest.fit(X_train, y_train)

错误：

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
/var/folders/hp/27sdwxlj1nqdg1ld2y84c43c0000gn/T/ipykernel_10792/3734698888.py in <module>
      3 
      4 bal_forest = BalancedRandomForestClassifier(n_estimators=100, random_state=1)
----> 5 bal_forest.fit(X_train, y_train)

~/opt/anaconda3/envs/mlenv/lib/python3.7/site-packages/imblearn/ensemble/_forest.py in fit(self, X, y, sample_weight)
    433 
    434         # Remap output
--> 435         _, self.n_features_ = X.shape
    436 
    437         y = np.atleast_1d(y)

AttributeError: can't set attribute

Answer 1

我解决了问题。

我不得不!pip安装不平衡-learn=0.9.0，然后是scikit-learn==1.0

其他版本删除.fit特性存在一个问题。

我从朱庇特笔记本公司转到谷歌公司，代码运行得很好。