mglearn KNeighborsClassifier不绘制边界

时间:2019-02-19 03:50:05

标签: python machine-learning knn supervised-learning

我目前正在使用从sklearn加载的乳腺癌数据集。我以这种方式将其变成了数据框:

data = load_breast_cancer()

#Load cancer dataset as dataframe
cancer = pd.DataFrame(np.c_[data['data'], data['target']], columns= np.append(data['feature_names'], ['target']))

我建立了KNN邻居分类模型并进行了测试。现在,我想使用mglearn来绘制决策边界,但我不断遇到此错误:   TypeError: unhashable type: 'slice'

这是我的代码:

import mglearn

fig, axes = plt.subplots(1, 3, figsize = (14, 5))

for n_neighbors, ax in zip([1, 7, 15], axes):
    clf = KNeighborsClassifier(n_neighbors = n_neighbors).fit(cancer, cancer["target"])
    mglearn.plots.plot_2d_separator(knn, cancer, fill = True, eps = 0.5, ax = ax, alpha = 0.4)
    mglearn.discrete_scatter(cancer.iloc[:, 0], cancer.iloc[:, 1], cancer["target"], ax = ax)
    ax.set_title("{} neighbor(s)".format(n_neighbors))
    ax.set_xlabel("feature 0")
    ax.set_ylabel("feature 1")
axes[0].legend(loc=3)

有什么建议吗?

0 个答案:

没有答案