def play_game(state_list, action_list, Qtable):
step = 0 # 合計ループ数
old_state = new_state = [] #- state
flag = 0
reset()
pt.init(0.5)
# メインループ
while(pt.alive()):
#########################
# ループごとの解析、操作をここに書く
if step % 180 == 0 and flag == 0:
new_board = pt.getBoard()
for i in range(len(new_board)):
for j in range(len(new_board[i])):
if new_board[i][j] > 0 and new_board[i][j] < 8: new_board[i][j] = 1
piece = pt.getPiece()
new_state = [new_board, piece]
if new_state != old_state and flag == 0:
action, value = egreedy.search_action(new_state)
if action == []:
action = egreedy.eGreedy(new_state)
old_state = new_state
flag = 1
if step % 10 == 0 and flag == 1:
flag = doaction(action, piece)
if step % 100 == 0: flag = 2
# print(new_state)
#########################
# 次のループへ
pt.loop()
if flag == 2:
print pt.param.flag['update']
pt.move('down')
if pt.param.flag['update']:
print "update"
pt.loop()
flag = 0
if pt.param.flag['gameover']: alive = flag = 0
step += 1
# 終了
pt.quit()
[8,0,0,0,0,0,0,0,0,0,0,8],
[8,0,0,0,0,0,0,0,0,0,0,8],
[8,0,0,0,0,0,0,0,0,0,0,8],
[8,0,0,0,0,0,0,0,0,0,0,8],
[8,8,8,8,8,8,8,8,8,8,8,8],
[8,8,8,8,8,8,8,8,8,8,8,8],
[8,8,8,8,8,8,8,8,8,8,8,8],
[8,8,8,8,8,8,8,8,8,8,8,8],
[8,8,8,8,8,8,8,8,8,8,8,8],
[8,8,8,8,8,8,8,8,8,8,8,8]]
# メインループ
while(pt.alive()):
#########################
# ループごとの解析、操作をここに書く
# if step%120==0:
# ans = pt.answer(m,[[1,1,1],[1,0,0],[1,0,0]], 8, -4, 2)
# pprint(ans)
#########################
# 次のループへ
pt.loop()
step += 1
# 終了
pt.quit()