qtable Namespace Reference

Functions
def	rendertrial (maxiter=100)

Variables
float	DECAY_RATE = 0.99
	env = DPendulum()
	— Environment More...
list	h_rwd = []
float	LEARNING_RATE = 0.85
int	NEPISODES = 500
	— Hyper paramaters More...
int	NSTEPS = 50
	NU = env.nu
	NX = env.nx
	Q = np.zeros([env.nx,env.nu])
	Qref = reward+DECAY_RATE*np.max(Q[x2,:])
	RANDOM_SEED = int((time.time()%10)*1000)
	— Random seed More...
	reward
float	rsum = 0.0
	u = np.argmax(Q[x,:] + np.random.randn(1,NU)/episode)
	x = env.reset()
	x2

Detailed Description

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

Function Documentation

def qtable.rendertrial

(

maxiter = 100

)

Roll-out from random state using greedy policy.

Definition at line 29 of file qtable.py.

Variable Documentation

float qtable.DECAY_RATE = 0.99

Definition at line 20 of file qtable.py.

qtable.env = DPendulum()

— Environment

Definition at line 23 of file qtable.py.

list qtable.h_rwd = []

Definition at line 39 of file qtable.py.

float qtable.LEARNING_RATE = 0.85

Definition at line 19 of file qtable.py.

int qtable.NEPISODES = 500

— Hyper paramaters

Definition at line 17 of file qtable.py.

int qtable.NSTEPS = 50

Definition at line 18 of file qtable.py.

qtable.NU = env.nu

Definition at line 25 of file qtable.py.

qtable.NX = env.nx

Definition at line 24 of file qtable.py.

qtable.Q = np.zeros([env.nx,env.nu])

Definition at line 27 of file qtable.py.

qtable.Qref = reward+DECAY_RATE*np.max(Q[x2,:])

Definition at line 49 of file qtable.py.

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

— Random seed

Definition at line 12 of file qtable.py.

qtable.reward

Definition at line 46 of file qtable.py.

float qtable.rsum = 0.0

Definition at line 43 of file qtable.py.

qtable.u = np.argmax(Q[x,:] + np.random.randn(1,NU)/episode)

Definition at line 45 of file qtable.py.

qtable.x = env.reset()

Definition at line 42 of file qtable.py.

qtable.x2

Definition at line 46 of file qtable.py.