sklearn-KN eighborsClassifier-ValueError：未知标签类型：'continuous'

Question

我不知道为什么这段代码不起作用？

    training_data_X    = np.array([ [1.2, 6.7, 2.7],  [2.3, 4.6, 2.2],  [0.3, 3.9, 0.8],  [2.1, 1.3, 4.3]  ])
    training_scores_Y  = np.array( [1.4, 9.2, 2.5, 2.2] )
    y_test  = np.array([ [1.2, 6.7, 2.7],  [7.6, 7.2, 0.2] ])

    knn = KNeighborsClassifier(n_neighbors=1)
    knn.fit(training_data_X, training_scores_Y)

    y_pred = knn.predict(y_test)
    print(y_pred)

我收到此错误：

ValueError: Unknown label type: 'continuous'

谢谢您的帮助。

Answer 1

• 您使用了KNN分类器，而您的问题是回归。

• 您可以KNN回归https://scikit-learn.org/stable/modules/generated/sklearn.neighbors.KNeighborsRegressor.html

Answer 2

看看https://www.kaggle.com/pratsiuk/valueerror-unknown-label-type-continuous

似乎您可以通过编码解决它：

lab_enc = preprocessing.LabelEncoder()
training_scores_encoded =lab_enc.fit_transform(training_scores_Y)