Output dimensionality after applying the Convolution
Given a Convolution Layer with 8 filters, a filter size of 6, a stride of 2, and a padding of 1. For an input feature map of 32 x 32 x 32, what is the output dimensionality after applying the Convolution Layer to the input?
Sample Answer
Here’s how to calculate the output dimensionality of the convolution layer:
- Effective Input Size:
Since padding is 1, for each dimension of the input (width and height), we add (2 * padding) to account for the padding added on either side. So, the effective input size becomes:
Width = Height = 32 + (2 * 1) = 34