知识卡片
卷积输出通道取决于核的数量
多个可训练滤波器产生多个输出通道;多通道输入时一个滤波器还跨输入通道聚合,不是只处理一张二维矩阵。堆叠层可学习层次表示,但每个通道不保证恰好对应可解释物体部件。
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
- · 深度学习模型
原书出处 · 1 处
- · 查看原书依据
Convolutional layers are the layers that, in large part, comprise CNNs. Convolutional layers consist of several kernels of the same dimension but different parameter values. Thus, a convolutional layer outputs several feature maps, where each feature map corresponds to the output of convolving one kernel with the input. As a result, the number of kernels specified to be in a convolutional layer is equal to
参考来源
- 位置:《Machine Learning Fundamentals: Concepts, Models, and Applications》第08章《深度学习模型》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Machine Learning Fundamentals: Concepts, Models, and Applications/_epub-src/OEBPS/xhtml/21_chapter08.xhtml)
- 结论依据:摘录中关于卷积输出通道取决于核的数量的定义、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 layers are the layers that, in large part, comprise CNNs. Convolutional layers consist of several kernels of the same dimension but different parameter values. Thus, a convolutional layer outputs several feature maps, where each feature map corresponds to the output of convolving one kernel with the input. As a result, the number of kernels specified to be in a convolutional layer is equal to