from introrl.utils import pickle_esp
start_time = time.time()
CR = CarRentalSimulation()
get_sim = Model(CR, build_initial_model=True)
get_sim.collect_transition_data(num_det_calls=50, num_stoic_calls=100000)
print('Total recorded actions Before:',
"{:,}".format(get_sim.total_num_action_data_points()))
CR.layout.s_hash_print()
get_sim.num_calls_layout_print(row_tickL=[c for c in ' First Location'],
const_col_w=True,
x_axis_label='Second Location',
none_str='*')
get_sim.min_num_calls_layout_print(
row_tickL=[c for c in ' First Location'],
const_col_w=True,
x_axis_label='Second Location',
none_str='*')
#get_sim.est_reward_error_layout_print(row_tickL=[c for c in ' First Location'], const_col_w=True,
# x_axis_label='Second Location', none_str='*')
#get_sim.define_statesD[(20,0)].summ_print()
#sys.exit() # <-------------------------------------
#get_sim.collect_transition_data( num_det_calls=10, num_stoic_calls=100 )
fname = os.path.split(__file__)[-1].split('.')[
0] # use file prefix for pickle file
print('Pickle File Name Prefix:', fname)
if not get_sim.read_pickle_file(fname):
get_sim.collect_transition_data(num_det_calls=10,
num_stoic_calls=10000)
#get_sim.collect_transition_data( num_det_calls=10, num_stoic_calls=10000 )
print('Total recorded actions Before:',
"{:,}".format(get_sim.total_num_action_data_points()))
BJ.layout.s_hash_print()
get_sim.num_calls_layout_print(
) #row_tickL=[c for c in ' Player Sum'], const_col_w=True,
#x_axis_label='Dealer Showing', none_str='*')
get_sim.min_num_calls_layout_print(
) # row_tickL=[c for c in ' Player Sum'], const_col_w=True,
#x_axis_label='Dealer Showing', none_str='*')
get_sim.est_reward_error_layout_print(
) #row_tickL=[c for c in ' Player Sum'], const_col_w=True,
#x_axis_label='Dealer Showing', none_str='*')
#sys.exit() # <-------------------------------------
get_sim.collect_transition_data(num_det_calls=10, num_stoic_calls=100)
print('Total recorded actions After:',
"{:,}".format(get_sim.total_num_action_data_points()))
from introrl.dp_funcs.dp_value_iter import dp_value_iteration
from introrl.environments.env_baseline import EnvBaseline
from introrl.agent_supt.model import Model
from introrl.utils import pickle_esp
from introrl.black_box_sims.blackjack_sim import BlackJackSimulation
start_time = time.time()
BJ = BlackJackSimulation()
get_sim = Model(BJ, build_initial_model=True)
get_sim.collect_transition_data(num_det_calls=50, num_stoic_calls=100000)
BJ.layout.s_hash_print()
get_sim.num_calls_layout_print()
get_sim.min_num_calls_layout_print()
print('got sim data')
print('_' * 55)
env = EnvBaseline(s_hash_rowL=BJ.s_hash_rowL,
x_axis_label=BJ.x_axis_label,
y_axis_label=BJ.y_axis_label)
get_sim.add_all_data_to_an_environment(env)
print('built environment')
print('_' * 55)
#env.summ_print()