def play_game(agent_1, eval_func_1, depth_1, agent_2, eval_func_2, depth_2, td_parameters, softmax_parameters, multilayer_parameters, board=losing_board.LosingBoard(no_kings=False)):
"""
Plays the specified game, printing progress, results and game duration to screen.
"""
# agent 1 evaluator
if eval_func_1 == evaluation.SoftmaxEval:
model = softmax.Softmax(*softmax_parameters)
model.train()
evaluator_1 = evaluation.SoftmaxEval(model).evaluate
elif eval_func_1 == evaluation.MultilayerEval:
model = multilayer.Multilayer(*multilayer_parameters)
model.train()
evaluator_1 = evaluation.MultilayerEval(model).evaluate
elif eval_func_1 == evaluation.TDTrainEval:
model = td_lambda.TDLeafLambda(*td_parameters)
model.train()
evaluator_1 = evaluation.TDTrainEval(model).evaluate
elif eval_func_1 is None:
evaluator_1 = None
else:
evaluator_1 = eval_func_1().evaluate
# agent 2 evaluator
if eval_func_2 == evaluation.SoftmaxEval:
model = softmax.Softmax(*softmax_parameters)
model.train()
evaluator_2 = evaluation.SoftmaxEval(model).evaluate
elif eval_func_2 == evaluation.MultilayerEval:
model = multilayer.Multilayer(*multilayer_parameters)
model.train()
evaluator_2 = evaluation.MultilayerEval(model).evaluate
elif eval_func_2 == evaluation.TDTrainEval:
model = td_lambda.TDLeafLambda(*td_parameters)
model.train()
evaluator_2 = evaluation.TDTrainEval(model).evaluate
elif eval_func_2 is None:
evaluator_2 = None
else:
evaluator_2 = eval_func_2().evaluate
# construct final agents, and gament_2(color=chess.BLACK, eval_func=evaluator_2, depth=depth_2, ant_eval_func=evaluator_1)
a1 = agent_1(color=chess.WHITE, eval_func=evaluator_1, ant_eval_func = evaluator_2, depth=depth_1)
a2 = agent_2(color=chess.BLACK, eval_func=evaluator_2, ant_eval_func = evaluator_1, depth=depth_2)
game_to_play = game.Game(board, a1, a2)
# begin
start = time.time()
game_to_play.play()
end = time.time()
print 'Game time:', (end - start)
def learnNums():
softmax_learn = softmax.Softmax()
trainDigits, trainLabels = softmax_learn.loadData('./train')
softmax_learn.train(trainDigits, trainLabels, maxIter=100) # 训练
return softmax_learn
def softmax(self, scores):
return sm.Softmax("").softmax_T(scores)