In matlab we use matconvnet's vl_nnconv with x, y. This is how it looks in matlab.
Example 1
x = randn(29, 29, 512)
y = randn(15, 15, 512)
[dzdx] = vl_nnconv(x, y)
Example 2
x = randn(29, 29, 512)
y = randn(15, 15, 512)
[dzdx] = vl_nnpool(x, y)
Assume that in Example 2 x is a matrix of size (29, 29, 512) and y is a matrix of size (15, 15, 512). The result is dzdx, of shape (29, 29, 512).
I am interested in exactly the same thing but in python, pytorch or keras precisely. I am trying to use just a simple numpy array (or a pytorch tensor/variable) for my equivalent of x and y (and weights). Is there anyway that can be achieved? Thanks
