sklearn-模型评价

import numpy as np

from sklearn.metrics import accuracy_score

y_pred=[0,2,1,3]
y_true=[0,1,2,3]


# 精度，预测=真实的数量／n
accuracy_res=accuracy_score(y_true,y_pred)
print(accuracy_res)

# kappa  介于-1到1之间，0.8以上的分数被认为一致性比较好

from sklearn.metrics import  cohen_kappa_score

kappa_res=cohen_kappa_score(y_true,y_pred)
print(kappa_res)


# 混淆矩阵

y_pred=[0,0,1,1]
y_true=[0,1,1,1]


from sklearn.metrics import confusion_matrix
confuse_res=confusion_matrix(y_true,y_pred)
print(confuse_res)

tn,fp,fn,tp=confusion_matrix(y_true,y_pred).ravel()
print(tn,fp,fn,tp)

'''   
        实际
预测     0  1
0     [1 , 0]
1     [1 , 2]

第一个1 表示实际为0预测为0的个数为1个

第二个1 表示实际为0预测为1的个数为1个

2表示 实际为1预测为1的个数为2个 

'''

# 分类报告

from sklearn.metrics import classification_report
print(classification_report(y_true,y_pred))



#hamming损失 1-accuracy

from sklearn.metrics import  hamming_loss

print(hamming_loss(y_true,y_pred))


#jaccard系数


from sklearn.metrics import  jaccard_similarity_score
print(jaccard_similarity_score(y_true,y_pred))
print(accuracy_score(y_true,y_pred))

# recall / precision / f


from sklearn.metrics import precision_score,recall_score,f1_score

print(precision_score(y_true,y_pred))
print(recall_score(y_true,y_pred))
print(f1_score(y_true,y_pred))