def initialize():
np.random.seed(1)
torch.manual_seed(1)
return EpsGreedy(Parameter(1)), GridWorld(2, 2, start=(0, 0), goal=(1, 1)),\
Gym(name='MountainCar-v0', horizon=np.inf, gamma=1.)
def test_grid_world():
np.random.seed(1)
mdp = GridWorld(start=(0, 0), goal=(2, 2), height=3, width=3)
mdp.reset()
mdp.render()
for i in range(10):
ns, r, ab, _ = mdp.step([np.random.randint(mdp.info.action_space.n)])
mdp.render()
assert ns == 5
np.random.seed(1)
mdp = GridWorldVanHasselt()
mdp.reset()
for i in range(10):
ns, r, ab, _ = mdp.step([np.random.randint(mdp.info.action_space.n)])
assert ns == 2
np.random.seed(5)
mdp = generate_grid_world('tests/environments/grid.txt', .9, 1, -1)
mdp.reset()
for i in range(10):
ns, r, ab, _ = mdp.step([np.random.randint(mdp.info.action_space.n)])
assert ns == 4