python实现汉明损失(hamming_loss)、0-1 损失(zero_one_loss)、覆盖误差(coverage_error)、排名损失(label_ranking)以及平均精度损失

python实现汉明损失(hamming_loss)、0-1 损失(zero_one_loss)、覆盖误差(coverage_error)、排名损失(label_ranking)以及平均精度损失

# -*- coding: utf-8 -*-
"""
Created on Thu Jul 18 15:26:34 2019

@author: muli
"""

import numpy as np

from sklearn.metrics import hamming_loss
from sklearn.metrics import zero_one_loss
from sklearn.metrics import coverage_error
from sklearn.metrics import label_ranking_loss
from sklearn.metrics import average_precision_score


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

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

h=hamming_loss(y_true, y_pred)  
print("汉明损失：",h)

z=zero_one_loss(y_true, y_pred)
print("0-1 损失：",z)

c=coverage_error(y_true, y_pred)-1  # 减 1原因：看第2个参考链接
print("覆盖误差：",c)

r=label_ranking_loss(y_true, y_pred)
print("排名损失：",r)

a=average_precision_score(y_true, y_pred) 
print("平均精度损失：",a)

参考链接：http://www.doc88.com/p-4932506209360.html
参考链接：https://sklearn.apachecn.org/#/docs/32?id=_33-模型评估--量化预测的质量