知识卡片
「模型」的三层含义:架构、定义与实例
「model」常被含糊地指三个相关却不同的概念。模型架构是本应用的学习策略,含模型族选择(DNN 还是随机森林)与层数、树数等结构决定。模型定义(configured model)是配置加训练环境、以及所训练数据的类型与定义:全量特征、全部超参数、随机种子及一切对复现重要的要素,可视为整个训练环境的闭包。训练模型则是某时刻在特定数据上训练出的具体快照,含一组权重与阈值。业界多把三者统称「模型」,本书也未必处处严格区分。
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
- Basic Introduction to Models · 查看原书依据
Model Architecture Versus Model Definition Versus Trained Model The term model is often used imprecisely to refer to three separate, but related, concepts: Model architecture The general strategy that we are using to learn in this application. This can include both the choice of model family, such as deep neural network (DNN) or random forest, and structural choices such as the number of layers in the DNN or the number of tree
参考来源
- 位置:《Reliable Machine Learning》第03章《Basic Introduction to Models》(源文件:extracted/Reliable Machine Learning/_epub-src/OEBPS/ch03.xhtml)
- 结论依据:原文明确阐述该主题,并给出定义、实践或反例。
- 原始内容:Model Architecture Versus Model Definition Versus Trained Model The term model is often used imprecisely to refer to three separate, but related, concepts: Model architecture The general strategy that we are using to learn in this application. This can include both the choice of model family, such as deep neural network (DNN) or random forest, and structural choices such as the number of layers in the DNN or the number of tree