Mobile Deep Learning with TensorFlow Lite, ML Kit and Flutter

Allowing moves according to the game's rules

To establish the validity of the moves made by the players, be it man or machine, we need to have a mechanism in place that continuously either only generates valid moves, in the case of the machine, or keeps validating any human player input. Let's get started:

1. One such instance can be found in the _selfplay(self, state, args) function in the util/generator.py file, as shown in the following code:

turn = 0hard_random_turn = args['hard_random'] if 'hard_random' in args else 0soft_random_turn = (args['soft_random'] if 'soft_random' in args else 30) + hard_random_turnhistory = []

First, we set the toggle for moves to 0, indicating that no moves have been made at the start of the game. We also ...