Mobile Deep Learning with TensorFlow Lite, ML Kit and Flutter

Creating an SRGAN model for super-resolution

First, we'll begin by working on the train.py file:

1. Let's start by importing the necessary modules into the project:

import osfrom data import DIV2Kfrom model.srgan import generator, discriminatorfrom train import SrganTrainer, SrganGeneratorTrainer

The preceding imports bring in some ready-made classes, such as SrganTrainer, SrganGeneratorTrainer, and so on. We will discuss each of them in detail after we've finished working on this file.

2. Now, let's create a directory for the weights. We shall use this directory to store intermediate models as well:

weights_dir = 'weights'weights_file = lambda filename: os.path.join(weights_dir, filename)os.makedirs(weights_dir, exist_ok=True)

3. Next, we'll ...