知识卡片
模型概率不是个人身份置信证明
分类输出的概率反映模型在指定类别和训练目标下的分布,并未自动校准到现实人群。它不能据姓氏确定某个人国籍,top几项也未必加到1;应将任务标签解释与真实身份区分。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
原书出处 · 1 处
- · 查看原书依据
This means that the model is only 31% confident that Rochdi is Algerian. These probabilities (likelihoods) can be used to explain how confident your model is to your boss, teammates, or even users.
参考来源
- 位置:《Natural Language Processing in Action, Second Edition》第08章《循环网络与长短期记忆》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Natural Language Processing in Action, Second Edition/_epub-src/OEBPS/Text/chapter-8.html)
- 结论依据:摘录中关于模型概率不是个人身份置信证明的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。
- 原始内容:This means that the model is only 31% confident that Rochdi is Algerian. These probabilities (likelihoods) can be used to explain how confident your model is to your boss, teammates, or even users.