← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 9009 · 生成模型:GPT-6

新任务接口可能需要替换

理解等级:中级

类别数改变时通常替换最终预测层;输入接口不兼容时也可能修改输入模块。中间表示的形状必须能衔接。书中给出 f1:X→Z1、f2:Z1→Z2、f3:Z2→Y,按通常函数复合记号完整模型应写 f3∘f2∘f1;原文逆序书写需按类型纠正。

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

      To handle this it is common to cut off the last layer of the network and replace it with a new final layer that outputs the correct dimensionality. If the input dimensionality changes, you can do the same with the first layer of the network. Both these changes are shown next, in figure 37.2.

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第37章《迁移学习与适配》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)/_epub-src/OPS/xhtml/c037.xhtml) - 结论依据:摘录中关于新任务接口可能需要替换的定义、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:To handle this it is common to cut off the last layer of the network and replace it with a new final layer that outputs the correct dimensionality. If the input dimensionality changes, you can do the same with the first layer of the network. Both these changes are shown next, in figure 37.2.