估计阅读时长: 8 分钟https://github.com/rsharp-lang/R-sharp 对于0,1两类分类问题,一些分类器得到的结果往往不是0,1这样的标签。如神经网络得到诸如0.5,0.8这样的分类结果。这时,我们人为取一个阈值,比如0.4,那么小于0.4的归为0类,大于等于0.4的归为1类,可以得到一个分类结果。同样,这个阈值我们可以取0.1或0.2等等。 Order by Date Name Attachments ROC • 221 kB • 335 click 2021年6月28日Roccurves • […]
Hello There. I found your website through the usage of msn. This is a very smartly written article. I will…
[…] […]
[…] […]
[…] 《为大语言模型运行添加工具调用》 […]
[…] 《从头创建一个DeepSeek客户端》 […]