def q3():
source = (0, 0)
target = (10, 10)
my_grid = environment.Environment(source, target)
random_policy = randomPolicy.RandomPolicy()
simulation_object_random = Simulation.Simulation(my_grid, random_policy)
simulation_object_random.simulate()
return
def q4():
source = (0, 0)
# fixed target
target = (10, 10)
my_grid = environment.Environment(source, target)
worse_policy = worsePolicy.WorsePolicy()
random_policy = randomPolicy.RandomPolicy()
better_policy = betterPolicy.BetterPolicy(target)
simulation_object_random = Simulation.Simulation(my_grid, random_policy)
simulation_object_random.simulate()
simulation_object_worse = Simulation.Simulation(my_grid, worse_policy)
simulation_object_worse.simulate()
simulation_object_better = Simulation.Simulation(my_grid, better_policy)
simulation_object_better.simulate()
return
def q2():
source = (0, 0)
target = (10, 10)
my_grid = environment.Environment(source, target)
curr_state = State.State(source[0], source[1], my_grid)
manual_policy = manualPolicy.ManualPolicy()
simulation_object_manual = Simulation.Simulation(my_grid, manual_policy)
while True:
action = manual_policy.get_action(curr_state)
action_object = Action.Action(action)
curr_state, reward = simulation_object_manual.transition(
curr_state, action_object)
print("Next state : ", curr_state.x, " ", curr_state.y)
print("Reward obtained : ", reward)
return
def q1():
source = (0, 0)
target = (10, 10)
my_grid = environment.Environment(source, target)
print("Enter current state : ")
x = int(input())
y = int(input())
curr_state = State.State(x, y, my_grid)
print("Enter desired action : ")
action = input()
action_object = Action.Action(action)
random_policy = randomPolicy.RandomPolicy()
simulation_object_random = Simulation.Simulation(my_grid, random_policy)
curr_state, reward = simulation_object_random.transition(
curr_state, action_object)
print("Next state : ", curr_state.x, " ", curr_state.y)
print("Reward obtained : ", reward)
return
def q5():
source = (0, 0)
# random target
target = (randint(0, 10), randint(0, 10))
my_grid = environment.Environment(source, target)
worse_policy = worsePolicy.WorsePolicy()
random_policy = randomPolicy.RandomPolicy()
better_policy = betterPolicy.BetterPolicy(target)
learned_policy = learnedPolicy.LearnedPolicy()
simulation_object_random = Simulation.Simulation(my_grid, random_policy)
simulation_object_random.simulate()
simulation_object_learned = Simulation.Simulation(my_grid, learned_policy)
simulation_object_learned.simulate()
simulation_object_worse = Simulation.Simulation(my_grid, worse_policy)
simulation_object_worse.simulate()
simulation_object_learned = Simulation.Simulation(my_grid, better_policy)
simulation_object_learned.simulate()
learned_policy.print_lookup()
return