Mobile Deep Learning with TensorFlow Lite, ML Kit and Flutter

Creating a Flask server script

In this section, we shall work on the flask_app.py file, which will be running on the cloud virtual machine as a server. Let's get started:

1. First, we'll make the necessary imports to the file:

from flask import Flask, request, jsonify, send_fileimport osimport timefrom matplotlib.image import imsavefrom model.srgan import generatorfrom model import resolve_single

2. Now, we'll define the weights directory and load the generator weights into the file:

weights_dir = 'weights'weights_file = lambda filename: os.path.join(weights_dir, filename)gan_generator = generator()gan_generator.load_weights(weights_file('gan_generator.h5'))

3. Next, we'll instantiate the Flask app using the following line of code:

app = ...