模型评估：分类器和回归器的评估

分类器的评估

混淆矩阵

from sklearn.metrics import confusion_matrix

y_true = [2, 0, 2, 2, 0, 1]

y_pred = [0, 0, 2, 2, 0, 2]

confusion_matrix(y_true, y_pred)

>>>array([ [2, 0, 0],

         [0, 0, 1],

         [1, 0, 2] ])

个人理解：每一行的行索引代表一个类别，每一行代表一个类别被预测分到任意类别的个数

准确率

from sklearn.metrics import accuracy_score

y_pred = [0, 2, 1, 3]

y_true = [0, 1, 2, 3]

accuracy_score(y_true, y_pred)

>>>0.5

accuracy_score(y_true, y_pred, normalize=False)

>>>2

Jaccard相似度

from sklearn.metrics import jaccard_similarity_score

y_pred = [0, 2, 1, 3]

y_true = [0, 1, 2, 3]

jaccard_similarity_score(y_true, y_pred)

>>>0.5

jaccard_similarity_score(y_true, y_pred, normalize=False)

>>>2

分类报告

该classification_report函数构建一个显示主分类指标的文本报告。

from sklearn.metrics import classification_report

y_true = [0, 1, 2, 2, 0]

y_pred = [0, 0, 2, 1, 0]

target_names = ['class 0', 'class 1', 'class 2']

print(classification_report(y_true, y_pred, target_names=target_names))

>>> 

              precision     recall    f1-score   support

    class 0       0.67      1.00      0.80         2

    class 1       0.00      0.00      0.00         1

    class 2       1.00      0.50      0.67         2

avg / total       0.67      0.60      0.59         5

roc_auc_score

from sklearn.metrics import roc_auc_score

y_true = np.array([0, 0, 1, 1])

y_scores = np.array([0.1, 0.4, 0.35, 0.8])

roc_auc_score(y_true, y_scores)

>>>0.75

roc_curve

import numpy as np

from sklearn import metrics

y = np.array([1, 1, 2, 2])

scores = np.array([0.1, 0.4, 0.35, 0.8])

fpr, tpr, thresholds = metrics.roc_curve(y, scores, pos_label=2)

#打印

fpr

>>>array([0. , 0. , 0.5, 0.5, 1. ])

tpr

>>>array([0. , 0.5, 0.5, 1. , 1. ])

thresholds

>>>array([1.8 , 0.8 , 0.4 , 0.35, 0.1 ])

其他

from sklearn.metrics import f1_score

from sklearn.metrics import precision_score

回归器的评估

MSE（mean square error，均方误差）

from sklearn.metrics import mean_squared_error

MAE（平均绝对误差）

from sklearn.metrics import mean_absolute_error

y_true = [3, -0.5, 2, 7]

y_pred = [2.5, 0.0, 2, 8]

mean_absolute_error(y_true, y_pred)

>>>0.5

Quantiles of Errors （中间绝对误差）

为了改进RMSE的缺点，提高评价指标的鲁棒性，使用误差的分位数来代替，如中位数来代替平均数。假设100个数，最大的数再怎么改变，中位数也不会变，因此其对异常点具有鲁棒性。

from sklearn.metrics import median_absolute_error

y_true = [3, -0.5, 2, 7]

y_pred = [2.5, 0.0, 2, 8]

median_absolute_error(y_true, y_pred)

>>>0.5

R-square(决定系数)

 

from sklearn.metrics import r2_score

y_true = [3, -0.5, 2, 7]

y_pred = [2.5, 0.0, 2, 8]

r2_score(y_true, y_pred) 

>>>0.948