# both agents interact (times):
interactions = 300
#
payouts, moves, states = morl_interact_multiple_episodic(
chebyagent,
problem,
interactions,
max_episode_length=300,
discounted_eps=False)
# print("TEST(cheby): interactions made: \nP: "+str(payouts[:])+",\n M: " + str(moves[:]) + ",\n S: " +
# str(states[:]) + '\n')
#, moves, states = morl_interact_multiple_average_episodic(chebyagent, problem, 10, 500)
# time = problem.time_token
chebyagent._epsilon = 0.9
payouts, moves2, states = morl_interact_multiple_episodic(
chebyagent, problem, 1, 300)
velocity = problem.get_velocities(states)
states = problem.create_plottable_states(states)
plot_hypervolume([chebyagent], problem)
forward_acc = []
backward_acc = []
nothin = []
for i in xrange(len(moves)):
counter = list(moves[i])
nothin.append(counter.count(0))
forward_acc.append(counter.count(1))
backward_acc.append(counter.count(2))
x = np.arange(len(nothin))
if show_trend:
# agent interact (times):
interactions = 200
#
payouts, moves, states = morl_interact_multiple_episodic(
chebyagent,
problem,
interactions,
max_episode_length=300,
discounted_eps=False)
# print("TEST(cheby): interactions made: \nP: "+str(payouts[:])+",\n M: " + str(moves[:]) + ",\n S: " +
# str(states[:]) + '\n')
#, moves, states = morl_interact_multiple_average_episodic(chebyagent, problem, 10, 500)
# time = problem.time_token
chebyagent._epsilon = 1.0
payouts, moves2, states = morl_interact_multiple_episodic(
chebyagent, problem, 1, 300)
velocity = problem.get_velocities(states)
states = problem.create_plottable_states(states)
plot_hypervolume([chebyagent], problem)
forward_acc = []
backward_acc = []
nothin = []
for i in xrange(len(moves)):
counter = list(moves[i])
nothin.append(counter.count(0))
forward_acc.append(counter.count(1))
backward_acc.append(counter.count(2))
x = np.arange(len(nothin))
if show_trend: