[Machine learning algorithm and learning _3_ API Code articles] 2.9 Performance Evaluation Model

1. The classification model assessment

sklearn There are three ways to assess the quality of a predictive model,

1) Each module has a score model method;

2) cross-validation module assessment tools;

3) metrics evaluation module has some functions.

Here are metrics module API, others see Resources.

# Of test set predicted 
y_predict = clf.predict (X_test) 

# ---- output classification accuracy errors and 
Print ( ' number of samples misclassified:% D ' !% (= Android.permission.FACTOR. Y_predict) .sum () + ' a ' )
 from sklearn.metrics Import accuracy_score
 Print ( " accuracy:% .2f ' % accuracy_score (android.permission.FACTOR., y_predict))
 Print () 

# ---- classification result is output confusion matrix 
DEF my_confusion_matrix (y_true, y_pred):
     from sklearn.metrics Import confusion_matrix 
    Labels = list(set(y_true))
    conf_mat = confusion_matrix(y_true, y_pred, labels = labels)
    print("confusion_matrix(left labels: y_true, up labels: y_pred)")
    print("labels",'\t',end='')
    for i in range(len(labels)):
        print(labels[i],'\t',end='')
    print()
    for i in range(len(conf_mat)):
        print(i,'\t',end='')
        for j in range(len(conf_mat[i])):
            print(conf_mat[i][j],'\t',end='')
        print()
    
my_confusion_matrix(y_test,y_predict)
print()

#----输出分类报告
from sklearn.metrics import classification_report
print("classification_report(left: labels):")
print(classification_report(y_test, y_predict))

View Code

 

 

References:

Machine learning and performance indicators used in the model evaluation sklearn

https://blog.csdn.net/sinat_36972314/article/details/82734370