qnet Namespace Reference

Classes
class	QValueNetwork
	— Q-value networks More...

Functions
def	disturb (u, i)
def	onehot (ix, n=NX)
def	rendertrial (maxiter=100)

Variables
float	DECAY_RATE = 0.99
	env = DPendulum()
	— Environment More...
	feed_dict
list	h_rwd = []
	— History of search More...
float	LEARNING_RATE = 0.1
int	NEPISODES = 500
	— Hyper paramaters More...
int	NSTEPS = 50
	NU = env.nu
	NX = env.nx
	optim
	Q2 = sess.run(qvalue.qvalue,feed_dict={ qvalue.x: onehot(x2) })
	Qref = sess.run(qvalue.qvalue,feed_dict={ qvalue.x: onehot(x ) })
	qvalue = QValueNetwork()
	RANDOM_SEED = int((time.time()%10)*1000)
	— Random seed More...
	reward
float	rsum = 0.0
	sess = tf.InteractiveSession()
	u = sess.run(qvalue.u,feed_dict={ qvalue.x: onehot(x) })[0]
	x = env.reset()
	— Training More...
	x2

Detailed Description

Example of Q-table learning with a simple discretized 1-pendulum environment using a linear Q network.

Function Documentation

def qnet.disturb	(		u,
			i
	)

Definition at line 58 of file qnet.py.

def qnet.onehot	(		ix,
			n = NX
	)

Return a vector which is 0 everywhere except index <i> set to 1.

Definition at line 54 of file qnet.py.

def qnet.rendertrial

(

maxiter = 100

)

Definition at line 62 of file qnet.py.

Variable Documentation

float qnet.DECAY_RATE = 0.99

Definition at line 23 of file qnet.py.

qnet.env = DPendulum()

— Environment

Definition at line 26 of file qnet.py.

qnet.feed_dict

Definition at line 90 of file qnet.py.

list qnet.h_rwd = []

— History of search

Definition at line 72 of file qnet.py.

float qnet.LEARNING_RATE = 0.1

Definition at line 22 of file qnet.py.

int qnet.NEPISODES = 500

— Hyper paramaters

Definition at line 20 of file qnet.py.

int qnet.NSTEPS = 50

Definition at line 21 of file qnet.py.

qnet.NU = env.nu

Definition at line 28 of file qnet.py.

qnet.NX = env.nx

Definition at line 27 of file qnet.py.

qnet.optim

Definition at line 90 of file qnet.py.

qnet.Q2 = sess.run(qvalue.qvalue,feed_dict={ qvalue.x: onehot(x2) })

Definition at line 85 of file qnet.py.

qnet.Qref = sess.run(qvalue.qvalue,feed_dict={ qvalue.x: onehot(x ) })

Definition at line 86 of file qnet.py.

qnet.qvalue = QValueNetwork()

Definition at line 50 of file qnet.py.

qnet.RANDOM_SEED = int((time.time()%10)*1000)

— Random seed

Definition at line 14 of file qnet.py.

qnet.reward

Definition at line 82 of file qnet.py.

float qnet.rsum = 0.0

Definition at line 77 of file qnet.py.

qnet.sess = tf.InteractiveSession()

Definition at line 51 of file qnet.py.

qnet.u = sess.run(qvalue.u,feed_dict={ qvalue.x: onehot(x) })[0]

Definition at line 80 of file qnet.py.

qnet.x = env.reset()

— Training

Definition at line 76 of file qnet.py.

qnet.x2

Definition at line 82 of file qnet.py.