使用 scikit-learn 执行 LDA 降维时出错

Question

I've got imported the Dataset from this URL in the pandas Dataframe called df : https://www.kaggle.com/jakeshbohaju/brain-tumor?select=Brain+Tumor.csv

但是，在运行线性判别分析时，我总是得到底部的错误。

from sklearn.discriminant_analysis import LinearDiscriminantAnalysis
X = df.drop(['label','Image'], axis=1)
y = df[['label']]

lda = LinearDiscriminantAnalysis(n_components=2)
X_r2 = lda.fit(X, y).transform(X)

错误：

ValueError                                Traceback (most recent call last) <ipython-input-41-f7a0f19db224> in <module>
     25 lda = LinearDiscriminantAnalysis(n_components=2)
     26 lda2 = LinearDiscriminantAnalysis(n_components=2)
---> 27 X_r3 = lda2.fit(X_train,y_train.values.ravel()).transform(X_train)
     28 X_r2 = lda.fit(X, y).transform(X)
     29 

~/miniforge3/envs/pyM1/lib/python3.8/site-packages/sklearn/discriminant_analysis.py in fit(self, X, y)
    537         else:
    538             if self.n_components > max_components:
--> 539                 raise ValueError(
    540                     "n_components cannot be larger than min(n_features, "
    541                     "n_classes - 1)."

ValueError: n_components cannot be larger than min(n_features, n_classes - 1).

Answer 1

解决方案

lda.fit(X, y)不返回任何内容，因此您不能在其上调用名为.transform()的方法。 API 仅允许您调用已定义的方法。 请参阅文档。

改成这样。 我还鼓励您花更多时间在文档上。

lda = LinearDiscriminantAnalysis(n_components=2)

# either use: lda.fit_transform(X, y)
X_r2 = lda.fit_transform(X, y)

## PREFERRED WAY
# or, use: lda.fit(X, y)
# followed by lda.transform(X)
lda.fit(X, y)
X_r2 = lda.transform(X)

参考

1. LinearDiscriminantAnalysis - 文档