We define a CNN model as fenn(X) = Softmax(FC (Conv2(MP (Relu (Convı (X)))))) The size of the input data X is 36 x 36 x

[answerhappygod]

We Define A Cnn Model As Fenn X Softmax Fc Conv2 Mp Relu Convi X The Size Of The Input Data X Is 36 X 36 X 1 (244.15 KiB) Viewed 23 times

We define a CNN model as fenn(X) = Softmax(FC (Conv2(MP (Relu (Convı (X)))))) The size of the input data X is 36 x 36 x 3; the first convolutional layer Convı includes 10 8 x 8 x 3 filters, stride=2, padding=1; Relui indicates the first Relu layer; MP, is a 2 x 2 max pooling layer, stride=2; the second convolutional layer Conv, includes 100 5 x 5 x 10 filters, stride=l, padding=0; FC indi- cates the fully connected layer, where there are 10 out- put neurons; Softmax denotes the Softmax activation function. The ground-truth label of X is denoted as t, and the loss function used for training this CNN model is denoted as (y,t). 1. Compute the feature map sizes after Reluz and Conv2 2. Calculate the number of parameters of this CNN model (hint: don't forget the bias parameter of in convolution and fully connection) 3. Plot the computational graph (CG) of the for- ward pass of this CNN model (hint: use 21, 22, 23, 24, 25, 26 denote the activated value after Convi, Relui, MP, Conv2, FC1, Softmax) 4. Based on the plotted CG, write down the formula- tions of back-propagation algorithm, including the forward and backward pass (Hint: for the forward pass, write down the process of how to get the value of loss function C(y,t); for the backward pass, write down the process of comput- ing the partial derivative of each parameter, like ac 端,盖)