From the course: Introduction to Generative Adversarial Networks (GANs)
Join today to access over 24,700 courses taught by industry experts.
Start training and compare results
From the course: Introduction to Generative Adversarial Networks (GANs)
Start training and compare results
- [Instructor] In the previous video, we gave some examples about how to set up the generative adversarial network training. We created the generator and the discriminator, data loader, loss function, and the optimizers. Now we'll go through the real engine and talk about how the training works. So, the training is essentially one large loop. It starts off with taking an epoch from a number so you can run the data through how many times you like. In this case, I've just put it as one, but a good idea would be to have it as 10 or a hundred or however patient you are. But just for an example, we'll just pass it through once. So that's how many times the entire dataset gets processed through the network. Then there's a smaller loop that actually samples each batch from the data loader and then pushes that through the network and then updates the weights. So let's go step-by-step. First, with the discriminator object, which is…
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.
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.