Mobile Deep Learning with TensorFlow Lite, ML Kit and Flutter

Deploying the script

Now, we will deploy the Flask server script. But before we do so, let's go through what this script does:

1. The first few lines of the script import the necessary modules for the script to work:

from flask import Flask, request, jsonifyimport osimport sysimport multiprocessing as mpfrom logging import getLoggerfrom chess_zero.agent.player_chess import ChessPlayerfrom chess_zero.config import Config, PlayWithHumanConfigfrom chess_zero.env.chess_env import ChessEnvfrom chess_zero.agent.model_chess import ChessModelfrom chess_zero.lib.model_helper import load_best_model_weightlogger = getLogger(__name__)

2. The rest of the code is put into the start() function, which is instantiated with a config object:

def start(config: ...