def test():
rows = 13
cols = 12
g = Grid(rows, cols, (int(rows / 2), 0), (int(cols / 2), cols - 1))
episode_len = 2 * (g.rows + g.cols)
episode_num = 50000
Q = g.q_learning(0.6, 0.1, 0.5, episode_len, episode_num)
pos_list = g.eval_control(Q, episode_len)
print(pos_list)
if pos_list[-1] != g.finish:
pos_list = [pos_list[0], pos_list[-1]]
ani = Draw.Animate(g)
ani.show(pos_list)
def test():
rows = 16
cols = 19
g = Grid(rows, cols, (0, 0), (rows-1, cols-1))
episode_len = (g.rows * g.cols)
episode_num = 100000
Q = g.q_learning(0.6, 0.1, 0.5, episode_len, episode_num)
pos_list = g.eval_control(Q, episode_len)
print(pos_list)
if pos_list[-1] != g.finish:
pos_list = [pos_list[0], pos_list[-1]]
ani = Draw.Animate(g)
ani.show(pos_list)
