def __init__(self, **args):
Agent.__init__(self, **args)
self.nb_finishFeatures = 4
self.nb_collisionFeatures = 4
self.nb_actions = 9
self.nb_features = 8
self.epsilon = 1
self.min_epsilon = 0.01
self.decay = .9999
self.discount = .9
self.finishAgent = DqnModule(nb_features=self.nb_finishFeatures,
featureExtractor=FeatureExtractor(
self.layout).getSimplestFeatures,
discount=self.discount)
self.collisionAgent = DqnModule(nb_features=self.nb_collisionFeatures,
featureExtractor=FeatureExtractor(
self.layout).getSimplestFeatures,
discount=0.8)
self.last_saved_num = -1
print '----------'
print '############ GmQAgent ############'
print 'Epsilon Decay = %s, Discount Factor = %.2f' % (self.decay,
self.discount)
print 'Reward: Finish+Time = %.2f, Collision = %.2f, FinishTimePenalty = %.2f, CollisionTimePenalty = %.2f' % \
(50*self.finishReward(TIME_STEP_PENALTY + FINISH_REWARD, 0), 50*self.collisionReward(TIME_STEP_PENALTY + COLLISION_PENALTY, 0), 50*self.finishReward(TIME_STEP_PENALTY, 0), 50*self.collisionReward(TIME_STEP_PENALTY, 0))
print '----------'
self.last_saved_num = -1
def __init__(self, **args):
Agent.__init__(self, **args)
self.nb_actions = 9
self.arbitrator_actions = 2
self.nb_finishFeatures = 4
self.nb_collisionFeatures = 4
self.nb_features = 4
self.outside_epsilon = 1.
self.min_epsilon = 0.01
self.outside_decay = .0
self.arbitratorDecay = .9995
self.finishDiscount = .9
self.collisionDiscount = .8
self.arbitratorDiscount = .9
self.finishAgent = DqnModule(nb_features=self.nb_finishFeatures,
featureExtractor=FeatureExtractor(
self.layout).getSimplestFeatures,
nb_actions=self.nb_actions,
discount=self.finishDiscount)
self.collisionAgent = DqnModule(nb_features=self.nb_collisionFeatures,
featureExtractor=FeatureExtractor(
self.layout).getSimplestFeatures,
nb_actions=self.nb_actions,
discount=self.collisionDiscount)
self.arbitrator = DDPGModule(nb_features=self.nb_features,
featureExtractor=FeatureExtractor(
self.layout).getSimplestFeatures,
nb_actions=self.arbitrator_actions,
discount=self.arbitratorDiscount,
decay=self.arbitratorDecay)
self.last_saved_num = -1
print '----------'
print '############ HierarchicalDDPGAgent ############'
print 'Outside Epsilon Decay = %f' % (self.outside_decay)
print 'FinishAgent: Discount Factor = %.2f' % (self.finishDiscount)
print 'CollisionAgent: Discount Factor = %.2f' % (
self.collisionDiscount)
print 'Arbitrator: Epsilon Decay = %f, Discount Factor = %.2f' % (
self.arbitrator.decay, self.arbitratorDiscount)
print 'Reward: Finish+Time = %.2f, Collision = %.2f, FinishTimePenalty = %.2f, CollisionTimePenalty = %.2f' % \
(50*self.finishReward(TIME_STEP_PENALTY + FINISH_REWARD, 0), 50*self.collisionReward(TIME_STEP_PENALTY + COLLISION_PENALTY, 0), 50*self.finishReward(TIME_STEP_PENALTY, 0), 50*self.collisionReward(TIME_STEP_PENALTY, 0))
print '----------'
def __init__(self, **args):
Agent.__init__(self, **args)
self.epsilon = 1.0
self.min_epsilon = 0.01
self.decay = 0.9999
self.nb_features = 8
self.nb_actions = 9
self.discount = .95
self.Agent = DqnModule(featureExtractor=FeatureExtractor(
self.layout).getCollisionFeatures,
nb_features=self.nb_features,
discount=self.discount)
print '----------'
print '############ CollisionAgent ############'
print 'Epsilon Decay = %s, Discount Factor = %.2f' % (self.decay,
self.discount)
print '----------'
self.last_saved_num = -1
def __init__(self, **args):
Agent.__init__(self, **args)
self.nb_actions = 9
self.arbitrator_actions = 2
self.nb_finishFeatures = 4
self.nb_collisionFeatures = 4
self.nb_features = 4
self.min_epsilon = 0.01
self.arbitrator_epsilon = 1.
self.arbitrator_decay = .9995
self.arbitratorDiscount = .9
self.dqnAgent = DqnModule(nb_features=self.nb_features,
featureExtractor=FeatureExtractor(
self.layout).getSimplestFeatures,
discount=self.arbitratorDiscount,
nb_actions=self.nb_actions)
self.finishAgent = DqnModule(nb_features=self.nb_finishFeatures,
featureExtractor=FeatureExtractor(
self.layout).getSimplestFeatures,
nb_actions=self.nb_actions)
self.collisionAgent = DqnModule(nb_features=self.nb_collisionFeatures,
featureExtractor=FeatureExtractor(
self.layout).getSimplestFeatures,
nb_actions=self.nb_actions)
self.arbitrator = DDPGModule(nb_features=self.nb_features,
featureExtractor=FeatureExtractor(
self.layout).getSimplestFeatures,
nb_actions=self.arbitrator_actions,
discount=self.arbitratorDiscount,
decay=self.arbitrator_decay)
self.last_saved_num = -1
print '----------'
print '############ SequentialDDPGAgent ############'
print 'Arbitrator Epsilon Decay = %f, Discount Factor = %.2f' % (
self.arbitrator.decay, self.arbitratorDiscount)
print '----------'
def __init__(self, extractor='IdentityExtractor', **args):
Agent.__init__(self, **args)
self.nb_features = 4
self.nb_finishFeatures = 4
self.nb_collisionFeatures = 4
self.nb_actions = 9
self.arbitrator_actions = 2
self.arbitratorEpsilon = 1
self.min_epsilon = 0.01
self.arbitratorDecay = .9995
self.finishDiscount = .9
self.collisionDiscount = .8
self.arbitratorDiscount = .9
self.finishAgent = DqnModule(nb_features=self.nb_finishFeatures,
featureExtractor=FeatureExtractor(
self.layout).getSimplestFeatures,
discount=self.finishDiscount,
nb_actions=self.nb_actions)
self.collisionAgent = DqnModule(nb_features=self.nb_collisionFeatures,
featureExtractor=FeatureExtractor(
self.layout).getSimplestFeatures,
discount=self.collisionDiscount,
nb_actions=self.nb_actions)
self.arbitrator = DqnModule(nb_features=self.nb_features,
featureExtractor=FeatureExtractor(
self.layout).getSimplestFeatures,
discount=self.arbitratorDiscount,
nb_actions=self.arbitrator_actions)
self.subModules = [self.finishAgent, self.collisionAgent]
self.last_saved_num = -1
self.finishAgent = self.loadModel('finishAgent_2_1999')
self.collisionAgent = self.loadModel('collisionAgent_2_1999')
print '----------'
print '############ SequentialArbiQAgent ############'
print 'Arbitrator: Decay = %f, Discount Factor = %.2f' % (
self.arbitratorDecay, self.arbitratorDiscount)
print '----------'
def __init__(self, **args):
Agent.__init__(self, **args)
self.actor = self.loadModel('actor_3_1000')
self.finishAgent = self.loadModel('')
self.collisionAgent = self.loadModel('')
def __init__(self, **args):
Agent.__init__(self, **args)
self.finishAgent = self.loadModel('finishAgent_2_1999')
self.collisionAgent = self.loadModel('collisionAgent_2_1999')