Missing nn.LeakyReLU in models.py

[PREPONDERANCE]

Thanks for the code!

In models.py line 104 and line 118, I saw two missing nn.LeakyReLU after convolution blocks. To my knowledge, the conventions write it that each convolution block (or linear block) should be followed by an activation function, a leaky relu in this case. Besides, I did not see any additional notes in the original paper stating this architecture.

Point me in the right direction if I was wrong. Thanks!