SVM算法中各标签对应精确率,召回率,f1score
1.精确率,召回率,f1score
主要参考博文https://www.cnblogs.com/pinard/p/5993450.html
2.用sklearn计算各标签对应的精确率,召回率,f1score,support(测试样本个数)
https://scikit-learn.org/stable/modules/classes.html#module-sklearn.metrics
classification_report()
precision_recall_fscore_support()
recall_score()
计算召回率
labels_recall_score = recall_score(y_true,y_pred,average=None)
precision_score()
f1_score()
3.各标签对应准确度实践
1)安装与测试sklearn库
2)把liblinear包导入linux环境下,并按照readme make生成相应文件
3)测试liblinear
4)把数据导入数据库