← 返回起始笔记位置

知识卡片

普通读书笔记卡 · 9234.c · 生成模型:GPT-6

训练验证测试承担不同责任

理解等级:中级

训练集用于拟合,验证集用于模型与超参数选择,最终测试集用于相对独立的性能评价。反复按测试结果选方案会破坏独立性;时间或群组数据还不能默认随机切分。

这张卡片看不懂?

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

      In the holdout method, the largest data set is randomly divided into three subsets: A training set is a subset of the data set that is used to build predictive models. The validation set is a subset of the data set used to assess the performance of the model built in the training phase. This set allows for fine-tuning the parameters of the model and

    参考来源

    - 位置:《Machine Learning Fundamentals: Concepts, Models, and Applications》第02章《机器学习基础》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Machine Learning Fundamentals: Concepts, Models, and Applications/_epub-src/OEBPS/xhtml/13_chapter02.xhtml) - 结论依据:摘录中关于训练验证测试承担不同责任的定义、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:In the holdout method, the largest data set is randomly divided into three subsets: A training set is a subset of the data set that is used to build predictive models. The validation set is a subset of the data set used to assess the performance of the model built in the training phase. This set allows for fine-tuning the parameters of the model and