In the add_log function it's logging the information of every player in the game with the star threshold. Shouldn't it only log the data for the player who won the game? The policy bot doesn't attack the enemy's general because 50% of the time the training data suggests not to.
In the
add_logfunction it's logging the information of every player in the game with the star threshold. Shouldn't it only log the data for the player who won the game? The policy bot doesn't attack the enemy's general because 50% of the time the training data suggests not to.