This is the step that brings it all together: training! In this recipe, you'll learn how to put all of these networks together and train your Pix2Pix network to do a style transfer.
Spot check! Make sure you have the following files in your working directory:
├── docker │ ├── build.sh │ ├── clean.sh │ └── Dockerfile ├── README.md ├── run.sh └── src | ├── generator.py | ├── discriminator.py | ├── gan.py | ├── train.py
Make sure you have the generator, discriminator, and GAN networks all built—otherwise, nothing in the training script will work!
This is how we start training our models - we need to create the right connections to each of the networks and inputs in the class instantiation,. build the training method that allows up to train this network, and finally understand the helper functions that allow us to make all of this code possible.