We are building a CNN network for image classification. The image dmension is 64*64*3. Here is the network. CONV1: 64 filters of size 3*3*x1, stride 1, padding p1, (output image dimension is same as input); CONV2: 32 filters of size 5*5*x2, stride 1, padding p2, (output image dimension is same as input); POOL1: filter size 2*2, stride 2, padding 0 (output dimension is D) a) Find, D, x1, x2, p1, p2. (10) b) How many parameters this model needs to learn for this