Convolutional Neural Networks¶

• no pad (valid): n, f -> n - f + 1. p is zero
• after pad (Same): n + p, f -> n + 2p - f + 1. $p = \frac{f - 1}{2}$ is the layer of the pad
• The value f is always odd.
• stride $s$. output dimention: floor of $\frac{n + 2p - f}{s} + 1$
• 2D image
• num_channel must match

input: 1 x 60 x 60, filter: 1 x 1 x 60,