[英]LIBSVM Evaluation with F-Score, Precision or Recall
I am new to Libsvm, and I just tried doing cross validation using grid.py and it gave me only accuracy evaluation. 我是Libsvm的新手,我只是尝试使用grid.py进行交叉验证,它只给了我准确性评估。
How do I make it show F-Score (preferred), or Precision and Recall? 如何使它显示F-Score(首选)或Precision and Recall?
Thanks! 谢谢!
The authors of libsvm
have provided an extension which enables you to use different scoring mechanisms during cross-validation, depending on your needs. libsvm
的作者提供了一个扩展 ,使您可以根据需要在交叉验证期间使用不同的评分机制。 Currently, it supports precision
, recall
, fscore
, bac
and auc
. 当前,它支持
precision
, recall
, fscore
, bac
和auc
。
Unfortunately, setting it up does involve a bit of manual work, so, if you need any help, please let me know. 不幸的是,设置它确实需要一些手动工作,因此,如果您需要任何帮助,请告诉我。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.