10-1 Adversarial Generative Network and DCGAN

The matrix above * the first row of the matrix = B00 a00*w00 + a01*w01 + a02*w02 + a10*w10 + a11*w11 + a12*w12 + a20*w20 + a21*w21 + a22*w22 
the matrix above* The second row
of the matrix = the matrix above B01 * the third row
of the matrix = the matrix above B10 * the fourth row of the matrix = B11

Model structure

1. Pooling layer is replaced with convolutional layer
2. Use strided covolutions on D
3. Use fractional-strided convolutions on G
4. Both G and D use batch normalization to help solve the problem of poor initialization, and the gradient is propagated to each layer
5. BN should not be used in the input and output layers
6. Remove the fully connected layer and use global pooling
7. Relu is used on G except for the output layer using tanh
8. Use LeakyReLU on D