知识卡片

普通读书笔记卡 · 9174.c · 生成模型:GPT-6

一维卷积可提取时间局部模式

理解等级:中级

把传感序列按时间与通道组织,可用1D卷积发现波峰等局部结构并用于分类。感受野限制能看到的时间范围,边界与是否因果还影响在线使用;换成时序输入不自动使模型拥有无限历史。

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

      Convolutional neural networks are not limited to 2D image data. They can also be applied to 1D time series. Here, the CNNs will find distinct patterns—for example, peaks in an accelerometer or gyroscope reading, which can then be used collectively to classify complex signals.

    参考来源

    - 位置:《Introduction to Autonomous Robots Mechanisms, Sensors, Actuators, and Algorithms (Nikolaus Correll, Bradley Hayes etc.)》第10章《人工神经网络》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Introduction to Autonomous Robots Mechanisms, Sensors, Actuators, and Algorithms (Nikolaus Correll, Bradley Hayes etc.)/_epub-src/OEBPS/xhtml/chapter_10.xhtml) - 结论依据:摘录中关于一维卷积可提取时间局部模式的定义、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:Convolutional neural networks are not limited to 2D image data. They can also be applied to 1D time series. Here, the CNNs will find distinct patterns—for example, peaks in an accelerometer or gyroscope reading, which can then be used collectively to classify complex signals.