RunEFR-CCE.py

# Copyright 2019 DeepMind Technologies Limited
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#      http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

"""Example use of the CFR algorithm on Kuhn Poker."""
import pandas as pd
import time
import sys


import EFR
from open_spiel.python.algorithms import exploitability
from open_spiel.python.algorithms import cfr
from open_spiel.python import policy
from multiprocessing import Pool

import StoreTabularPolicy
import pyspiel

FL_game = 'kuhn_poker'
FL_iterations = 2
FL_players = 3
FL_print_freq = 1
FL_Runname = "Test run"
try:
  FL_Runname = sys.argv[1]
except:
  pass
try:
  FL_iterations = int(sys.argv[2])
  print(FL_iterations)
except:
  pass




def main(deviation):
  header = ['Iteration', 'Distance to CCE']#, 'Average Payoff', 'Value compared to uniform']
  efr_data = []
  game = pyspiel.load_game(FL_game, {"players": FL_players})
  efr_solver = EFR.EFRSolver(game, deviation)
  strategies = []
  for i in range(FL_iterations):
    EFR_iteration_data = [i]
    efr_solver.evaluate_and_update_policy()

    if i % FL_print_freq == 0:
      strategies.append(policy.python_policy_to_pyspiel_policy(
          efr_solver.current_policy()))      
      corr_dev = pyspiel.uniform_correlation_device(strategies)
      cce_dist_info = pyspiel.cce_dist(game, corr_dev)
      print("EFR {} Iteration {} distance to coarse correlated {}".format(deviation, str(i), cce_dist_info.dist_value))

      EFR_iteration_data.append(cce_dist_info.dist_value)
    efr_data.append(EFR_iteration_data)
    print ("##########################################")


  EFR_data = pd.DataFrame(efr_data, columns=header)
  EFR_data.to_csv(deviation+ " "+ FL_Runname+'EFR_CCE_3player.csv', index=False)



with Pool(processes=8) as pool:
  deviation_types = ["blind action", "informed action", "blind cf", "informed cf", "bps", "cfps", "csps", "tips", "bhv"]
  pool.map(main, deviation_types)