From the course: AI Workshop: Hands-on with GANs with Deep Convolutional Networks
Unlock this course with a free trial
Join today to access over 23,300 courses taught by industry experts.
Training a discriminator on good fakes
From the course: AI Workshop: Hands-on with GANs with Deep Convolutional Networks
Training a discriminator on good fakes
Next, we'll actually train the discriminator. And again, the parameters and the steps we follow are the same as in the previous demo. So I won't repeat the same information over and over again. Here is where we set up variables with the number of channels, the size of feature maps in the discriminator, and the learning rate for the optimizer. Next, we set up the function to initialize the weights of the discriminator based on the recommendation of the original DCGANs paper. Then, of course, we instantiate the class for the discriminator. This is the same discriminator network that we saw in the previous demo, and that we explored in a lot of detail, so I won't go through the explanations again. Notice that we have convolutional layers, batch normalization layers, and the LeakyReLU activation, and the last convolutional layer of the discriminator, of course, has the sigmoid activation. Now, after we've set up the discriminator, let's instantiate the discriminator and move it to the…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Contents
-
-
-
-
Types of convolutional layers4m 11s
-
(Locked)
Training data for discriminator bad fakes and real images4m 34s
-
(Locked)
Loading and transforming training image data4m 55s
-
(Locked)
Understanding the discriminator architecture7m 8s
-
(Locked)
Training a discriminator on bad fakes5m 58s
-
(Locked)
Training data for discriminator good fakes and real images3m 8s
-
(Locked)
Training a discriminator on good fakes4m 56s
-
-
-