The architecture of a Convolutional Neural Network is given as below. Assuming that a bias unit is added to each of the filters and fully connected layers, answer Q29 - Q37. Activation Shape Activation Size (32, 32, 3) (28, 28, 12) 3072 9408 Number of parameters (a) (b) (14, 14, 12) 2352 (C) (10, 10, 24) 2400 (d) Input layer Convolution 1 (filter size = 5, Stride = 1) Max Pooling (Window size = 2, Stride = 2) Convolution 2 (filter size = 5, Stride = 1) Max Pooling (Window size = 2, Stride = 2) Fully connected layer 3 Fully connected layer 4 Softmax (5,5, 24) 600 (e) (150, 1) 150 (f) (96, 1) 96 20 (20, 1) Total parameters (h) (i) Q29. What is the number of parameters of (a)? * Your answer Q30. What is the number of parameters of (b)? * Your answer Q31. What is the number of parameters of (c)? * Your answer Q32. What is the number of parameters of (d)? * Your answer Q33. What is the number of parameters of (e)? * Your answer Q34. What is the number of parameters of (f)? * Your answer Q35. What is the number of parameters of (g)? * Your answer Q36. What is the number of parameters of (h)? * Your answer Q37. What is the number of parameters of (i)? * Your