training_data = list() RANDOM_FACTOR=0 temp = 0 oldREWSUM = 0 while (1): clock.tick(60) for event in pygame.event.get(): #print(event) if event.type == pygame.QUIT:#for exiting the game Game.QUITGAME() if event.type == pygame.KEYDOWN: if event.key ==pygame.K_DOWN: MyFunctions.GRAPH_REWARDS(rewards_array_L, rewards_array_R, rounds_array) if np.random.binomial(1,EPSILON): #print (np.shape(STATE)) AVR = session.run(Q_R, feed_dict = {State_InR: [STATE]}) AVR = MyFunctions.ONE_HOT_ACTIONS(AVR) else: AVR = MyFunctions.RANDOM_ONE_HOT() if np.random.binomial(1,EPSILON): #print (np.shape(STATE)) AVL = session.run(Q_L, feed_dict = {State_InL: [STATE]}) AVL = MyFunctions.ONE_HOT_ACTIONS(AVL) else:
RRewSUM = RRewSUM + R #print([L,R,AVL,AVR, DATA]) training_data.append( [np.copy(OLD_STATE), L, R, AVL[:], AVR[:], np.copy(STATE)]) if (L != 0): NUM_ROUNDS_PLAYED = NUM_ROUNDS_PLAYED + 1 time_step = time_step + 1 for event in pygame.event.get(): #print(event) if event.type == pygame.KEYDOWN: if event.key == pygame.K_DOWN: MyFunctions.GRAPH_REWARDS(rewards_array) if event.type == pygame.QUIT: #for exiting the game results_file.close() Game.QUITGAME() if time_step % 5000 == 0: line = str(time_step) + ',' + str(LRewSUM) + ';' rewards_array.append(LRewSUM) print(line) results_file.write(line) temp = NUM_ROUNDS_PLAYED LRewSUM = 0 RRewSUM = 0 if time_step % 100000 == 0: