V={}
for s in g.state:
V[s]=0
grama = 0.9
n=0
while True:
n+=1
print("No of iteration :",n)
different = 0
for s in g.state:
if not g.is_terminated_state(s):
old_v = V[s]
new_v= 0
for a in policy[s]:
next_s = g.get_next_state(a,s)
new_v += g.reward.get(next_s,0) + (grama * V[next_s])
V[s] = new_v
different = max(different, abs(new_v-old_v))
if different<0.0003:
break
print_policy(policy,g)
