如何计算单个CNN层中的权重和偏差值？

Question

给出下图，如何计算参数数量：

CNN Layer

此特定层由4x4卷积和64个特征图组成；如何完成满足最初问题的计算？

Update - Full Architecture

Answer 1

过滤器大小包含N * kernel_size * kernel_size权重参数，每个通道一个，所以

N * kernel_size * kernel_size* n_channels然后N bais parmaters，所以该层的最终cacluation为n_params = N * kernel_size * kernel_size* n_channels + N

---

N : number of features

kernel_size : is conv2d shape (height and width)

n_channels : is the number of channels

n_params : is the total number of your parameters

Ex

n_params = 64 * (4 * 4) * 3 + 64 = 3136