'''    1.分类模型之混淆矩阵：            每一行和每一列分别对应样本输出中的每一个类别，行表示实际类别，列表示预测类别。                        A类别    B类别    C类别                A类别    5        0        0                B类别    0        6        0                C类别    0        0        7            上述矩阵即为理想的混淆矩阵。不理想的混淆矩阵如下：                        A类别    B类别    C类别                A类别    3        1        1                B类别    0        4        2                C类别    0        0        7            查准率 = 主对角线上的值 / 该值所在列的和            召回率 = 主对角线上的值 / 该值所在行的和    获取模型分类结果的混淆矩阵的相关API：            import sklearn.metrics as sm            sm.confusion_matrix(实际输出, 预测输出)->混淆矩阵    2.分类模型之分类报告：                sklearn.metrics提供了分类报告相关API，不仅可以得到混淆矩阵，还可以得到交叉验证查准率、召回率、f1得分的结果，                可以方便的分析出哪些样本是异常样本。            # 获取分类报告            cr = sm.classification_report(实际输出, 预测输出)'''import numpy as npimport matplotlib.pyplot as mpimport sklearn.naive_bayes as nbimport sklearn.model_selection as msimport sklearn.metrics as smdata = np.loadtxt('./ml_data/multiple1.txt', delimiter=',', unpack=False, dtype='f8')print(data.shape)x = np.array(data[:, :-1])y = np.array(data[:, -1])# 训练集和测试集的划分    使用训练集训练 再使用测试集测试，并绘制测试集样本图像train_x, test_x, train_y, test_y = ms.train_test_split(x, y, test_size=0.25, random_state=7)# 针对训练集，做5次交叉验证，若得分还不错再训练模型model = nb.GaussianNB()# 精确度score = ms.cross_val_score(model, train_x, train_y, cv=5, scoring='accuracy')print('accuracy score=', score)print('accuracy mean=', score.mean())# 查准率score = ms.cross_val_score(model, train_x, train_y, cv=5, scoring='precision_weighted')print('precision_weighted score=', score)print('precision_weighted mean=', score.mean())# 召回率score = ms.cross_val_score(model, train_x, train_y, cv=5, scoring='recall_weighted')print('recall_weighted score=', score)print('recall_weighted mean=', score.mean())# f1得分score = ms.cross_val_score(model, train_x, train_y, cv=5, scoring='f1_weighted')print('f1_weighted score=', score)print('f1_weighted mean=', score.mean())# 训练NB模型，完成分类业务model.fit(train_x, train_y)pred_test_y = model.predict(test_x)# 得到预测输出，可以与真实输出作比较，计算预测的精准度(预测正确的样本数/总测试样本数)ac = (test_y == pred_test_y).sum() / test_y.sizeprint('预测精准度 ac=', ac)# 获取混淆矩阵m = sm.confusion_matrix(test_y, pred_test_y)print('混淆矩阵为：', m, sep='\n')# 获取分类报告r = sm.classification_report(test_y, pred_test_y)print('分类报告为：', r, sep='\n')# 绘制分类边界线l, r = x[:, 0].min() - 1, x[:, 0].max() + 1b, t = x[:, 1].min() - 1, x[:, 1].max() + 1n = 500grid_x, grid_y = np.meshgrid(np.linspace(l, r, n), np.linspace(b, t, n))bg_x = np.column_stack((grid_x.ravel(), grid_y.ravel()))bg_y = model.predict(bg_x)grid_z = bg_y.reshape(grid_x.shape)# 画图mp.figure('NB Classification', facecolor='lightgray')mp.title('NB Classification', fontsize=16)mp.xlabel('X', fontsize=14)mp.ylabel('Y', fontsize=14)mp.tick_params(labelsize=10)mp.pcolormesh(grid_x, grid_y, grid_z, cmap='gray')mp.scatter(test_x[:, 0], test_x[:, 1], s=80, c=test_y, cmap='jet', label='Samples')mp.legend()mp.show()# 画出混淆矩阵mp.figure('Confusion Matrix')mp.xticks([])mp.yticks([])mp.imshow(m, cmap='gray')mp.show()输出结果：(400, 3)accuracy score= [1.         1.         1.         1.         0.98305085]accuracy mean= 0.9966101694915255precision_weighted score= [1.         1.         1.         1.         0.98411017]precision_weighted mean= 0.996822033898305recall_weighted score= [1.         1.         1.         1.         0.98305085]recall_weighted mean= 0.9966101694915255f1_weighted score= [1.         1.         1.         1.         0.98303199]f1_weighted mean= 0.9966063988235516预测精准度 ac= 0.99混淆矩阵为：[[22  0  0  0] [ 0 27  1  0] [ 0  0 25  0] [ 0  0  0 25]]分类报告为：              precision    recall  f1-score   support         0.0       1.00      1.00      1.00        22         1.0       1.00      0.96      0.98        28         2.0       0.96      1.00      0.98        25         3.0       1.00      1.00      1.00        25    accuracy                           0.99       100   macro avg       0.99      0.99      0.99       100weighted avg       0.99      0.99      0.99       100