"""
import numpy as np
import matplotlib.pyplot as plt
from gridWorld import standardGrid, negativeGrid
from iterativePolicyEvaluation import printPolicy, printValues
from monteCarloControl import argMax
from td0Prediction import randomAction
GAMMA = 0.9
ALPHA = 0.1
ACTIONS = {'U','D','L','R'}
if __name__ == '__main__':
grid = negativeGrid(stepCost = -0.1)
print ("Rewards:")
printValues(grid.rewards,grid)
Q = {}
states = grid.allStates()
for s in states:
Q[s] = {}
for a in ACTIONS:
Q[s][a] = 0
updateCountsSarsa = {}
Created on Mon Jan 7 11:37:45 2019
@author: user
"""
import numpy as np
from gridWorld import standardGrid, negativeGrid
from iterativePolicyEvaluation import printValues, printPolicy
EPSILON = 10e-4
GAMMA = 0.9
ACTIONS = ('U', 'D', 'L', 'R')
if __name__ == '__main__':
grid = negativeGrid()
policy = {}
for s in grid.actions.keys():
policy[s] = np.random.choice(ACTIONS)
print("Rewards:")
printValues(grid.rewards, grid)
print("Initial Policy:")
printPolicy(policy, grid)
V = {}
states = grid.allStates()
for s in states: