import easy21
import MC
from plot import *
import pickle
# Run Monte-Carlo Simulation
RERUN = True
if RERUN:
mc = MC.MonteCarlo21(N0=100.0, every_visit=False)
for i in range(1000000):
mc.run_episode()
# if i % 1000 == 0:
# print(mc._action_state_values[0:2,:,:])
# mc.plot_optimal_value_function()
f = open('mc.pickle', 'wb')
pickle.dump(mc, f)
else:
f = open('mc.pickle', 'rb')
mc = pickle.load(f)
plot_optimal_value_function(mc._action_state_values, mc._action_state_counts, show=True)
