def __init__(self, epsilon, lambda_=0.5):
self.feedSrv = rospy.Service('SuturoMlHeadNextAction', SuturoMlNextAction, self.nextActionCallback)
self.policyPringPub = rospy.Publisher('SuturoMlPolicy', String, queue_size=10, latch=True)
self.policy = []
# self.q = defaultdict(lambda : 10)
# self.actions = filter(lambda x: x[0].startswith('Const'), [(a,s) for a,s in vars(SuturoMlAction).iteritems()])
self.actions = ["GRAB-SIDE blue_handle",
"GRAB-SIDE red_cube",
"TURN",
"OPEN-GRIPPER",
"GRAB-TOP blue_handle",
"GRAB-TOP red_cube",
"PLACE-IN-ZONE"]
self.q = None
self.policyMaker = EpsilonGreedyPolicy(self.q, self.actions, epsilon)
# self.policyMaker = Haxx0rPolicy(self.actions)
self.learner = SarsaLambdaLearner(self.policyMaker, l=lambda_)
self.q = self.learner.get_q()
self.policyMaker.updateQ(self.q)
# self.policyMaker = ReverseGreedyPolicy(self.q, self.actions)
rospy.wait_for_service('json_prolog/simple_query')
self.prolog = Prolog()
print("SuturoMlHeadLearnerPolicyFeeder started.")
