def run_sigopt(box_scores, historical_games, historical_games_training_set, bet_info, sigopt_width=1, sigopt_depth=100):
historical_games_by_tuple = evaluator.get_historical_games_by_tuple(historical_games)
conn = sigopt.Connection()
experiment = create_sigopt_experiment(conn, sigopt_depth)
for _ in range(experiment.observation_budget):
tunable_param_lists = []
suggestion_ids = []
for worker_id in range(sigopt_width):
conn = sigopt.Connection()
suggestion = conn.experiments(experiment.id).suggestions().create()
suggestion_ids.append(suggestion.id)
moving_averages = (
suggestion.assignments['slow_ma'],
suggestion.assignments['fast_ma'],
)
transform_params = {
'type': EXP_TRANSFORM,
'exp_param': suggestion.assignments['exp_param'],
}
tunable_param_lists.append([
moving_averages,
transform_params,
suggestion.assignments['n_estimators'],
suggestion.assignments['min_samples_split'],
suggestion.assignments['min_samples_leaf'],
suggestion.assignments['bet_threshold'],
])
all_stats = read_data.generate_all_stats(box_scores)
winnings_list = runner(
historical_games_training_set,
historical_games_by_tuple,
bet_info,
all_stats,
tunable_param_lists,
)
for i, suggestion_id in enumerate(suggestion_ids):
conn.experiments(experiment.id).observations().create(
suggestion=suggestion_id,
value=winnings_list[i][0],
value_stddev=winnings_list[i][1],
)
print "Optimization done. View results at https://sigopt.com/experiment/{0}".format(experiment.id)
experiment_detail = conn.experiments(experiment.id).fetch()
best_observation = experiment_detail.progress.best_observation
if best_observation:
print "Best value found: {0} at {1}".format(best_observation.value, best_observation.assignments)
return experiment.id
def run_sigopt(
box_scores,
user_token,
client_token,
client_id,
historical_games,
historical_games_training_set,
bet_info,
sigopt_width=1,
sigopt_depth=100,
):
historical_games_by_tuple = evaluator.get_historical_games_by_tuple(historical_games)
conn = sigopt.interface.Connection(user_token=user_token, client_token=client_token)
experiment = create_sigopt_experiment(conn, client_id)
for _ in range(sigopt_depth):
tunable_param_lists = []
assignments = []
for worker_id in range(sigopt_width):
conn = sigopt.interface.Connection(user_token=user_token, client_token=client_token, worker_id=worker_id)
suggestion = conn.experiments(experiment.id).suggest().suggestion
assignments.append(suggestion.assignments)
moving_averages = (suggestion.assignments["slow_ma"], suggestion.assignments["fast_ma"])
transform_params = {"type": EXP_TRANSFORM, "exp_param": suggestion.assignments["exp_param"]}
tunable_param_lists.append(
[
moving_averages,
transform_params,
suggestion.assignments["n_estimators"],
suggestion.assignments["min_samples_split"],
suggestion.assignments["min_samples_leaf"],
suggestion.assignments["bet_threshold"],
]
)
all_stats = read_data.generate_all_stats(box_scores)
winnings_list = runner(
historical_games_training_set, historical_games_by_tuple, bet_info, all_stats, tunable_param_lists
)
for i, assignment in enumerate(assignments):
conn.experiments(experiment.id).report(
data={"assignments": assignment, "value": winnings_list[i][0], "value_stddev": winnings_list[i][1]}
)
print "Optimization done. View results at https://sigopt.com/experiment/{0}".format(experiment.id)
best_observation = conn.experiments(experiment.id).bestobservation().observation
if best_observation:
print "Best value found: {0} at {1}".format(best_observation.value, best_observation.assignments)
return experiment.id
def run_example(sigopt_width=1, sigopt_depth=100): boxscores_path = os.path.join(os.path.dirname(__file__), '../boxscores/all_boxscores.json') box_scores = read_data.read_box_scores(boxscores_path) historical_games = get_historical_games(box_scores) historical_games_training_set = get_historical_games(box_scores, max_date=SEASON_1314_END) bet_info_s15 = bet_reader.read_info() bet_info = bet_reader.transform_old_format(bet_info_s15) historical_games_by_tuple = evaluator.get_historical_games_by_tuple(historical_games) return run_sigopt(box_scores, historical_games, historical_games_training_set, bet_info, sigopt_width=sigopt_width, sigopt_depth=sigopt_depth)
def run_example(user_token, client_token, client_id, sigopt_width=1, sigopt_depth=100):
box_scores = read_data.read_box_scores("../boxscores/all_boxscores.json")
historical_games = get_historical_games(box_scores)
historical_games_training_set = get_historical_games(box_scores, max_date=SEASON_1314_END)
bet_info_s15 = bet_reader.read_info()
bet_info = bet_reader.transform_old_format(bet_info_s15)
historical_games_by_tuple = evaluator.get_historical_games_by_tuple(historical_games)
return run_sigopt(
box_scores,
user_token,
client_token,
client_id,
historical_games,
historical_games_training_set,
bet_info,
sigopt_width=sigopt_width,
sigopt_depth=sigopt_depth,
)
def run_example(sigopt_width=1, sigopt_depth=100):
boxscores_path = os.path.join(os.path.dirname(__file__),
'../boxscores/all_boxscores.json')
box_scores = read_data.read_box_scores(boxscores_path)
historical_games = get_historical_games(box_scores)
historical_games_training_set = get_historical_games(
box_scores, max_date=SEASON_1314_END)
bet_info_s15 = bet_reader.read_info()
bet_info = bet_reader.transform_old_format(bet_info_s15)
historical_games_by_tuple = evaluator.get_historical_games_by_tuple(
historical_games)
return run_sigopt(box_scores,
historical_games,
historical_games_training_set,
bet_info,
sigopt_width=sigopt_width,
sigopt_depth=sigopt_depth)
def run_example(user_token,
client_token,
client_id,
sigopt_width=1,
sigopt_depth=100):
box_scores = read_data.read_box_scores('../boxscores/all_boxscores.json')
historical_games = get_historical_games(box_scores)
historical_games_training_set = get_historical_games(
box_scores, max_date=SEASON_1314_END)
bet_info_s15 = bet_reader.read_info()
bet_info = bet_reader.transform_old_format(bet_info_s15)
historical_games_by_tuple = evaluator.get_historical_games_by_tuple(
historical_games)
return run_sigopt(box_scores,
user_token,
client_token,
client_id,
historical_games,
historical_games_training_set,
bet_info,
sigopt_width=sigopt_width,
sigopt_depth=sigopt_depth)
def run_sigopt(box_scores,
user_token,
client_token,
client_id,
historical_games,
historical_games_training_set,
bet_info,
sigopt_width=1,
sigopt_depth=100):
historical_games_by_tuple = evaluator.get_historical_games_by_tuple(
historical_games)
conn = sigopt.interface.Connection(
user_token=user_token,
client_token=client_token,
)
experiment = create_sigopt_experiment(conn, client_id)
for _ in range(sigopt_depth):
tunable_param_lists = []
assignments = []
for worker_id in range(sigopt_width):
conn = sigopt.interface.Connection(
user_token=user_token,
client_token=client_token,
worker_id=worker_id,
)
suggestion = conn.experiments(experiment.id).suggest().suggestion
assignments.append(suggestion.assignments)
moving_averages = (
suggestion.assignments['slow_ma'],
suggestion.assignments['fast_ma'],
)
transform_params = {
'type': EXP_TRANSFORM,
'exp_param': suggestion.assignments['exp_param'],
}
tunable_param_lists.append([
moving_averages,
transform_params,
suggestion.assignments['n_estimators'],
suggestion.assignments['min_samples_split'],
suggestion.assignments['min_samples_leaf'],
suggestion.assignments['bet_threshold'],
])
all_stats = read_data.generate_all_stats(box_scores)
winnings_list = runner(
historical_games_training_set,
historical_games_by_tuple,
bet_info,
all_stats,
tunable_param_lists,
)
for i, assignment in enumerate(assignments):
conn.experiments(experiment.id).report(
data={
'assignments': assignment,
'value': winnings_list[i][0],
'value_stddev': winnings_list[i][1],
})
print "Optimization done. View results at https://sigopt.com/experiment/{0}".format(
experiment.id)
best_observation = conn.experiments(
experiment.id).bestobservation().observation
if best_observation:
print "Best value found: {0} at {1}".format(
best_observation.value, best_observation.assignments)
return experiment.id
def run_sigopt(box_scores,
historical_games,
historical_games_training_set,
bet_info,
sigopt_width=1,
sigopt_depth=100):
historical_games_by_tuple = evaluator.get_historical_games_by_tuple(
historical_games)
conn = sigopt.Connection()
experiment = create_sigopt_experiment(conn, sigopt_depth)
for _ in range(experiment.observation_budget):
tunable_param_lists = []
suggestion_ids = []
for worker_id in range(sigopt_width):
conn = sigopt.Connection()
suggestion = conn.experiments(experiment.id).suggestions().create()
suggestion_ids.append(suggestion.id)
moving_averages = (
suggestion.assignments['slow_ma'],
suggestion.assignments['fast_ma'],
)
transform_params = {
'type': EXP_TRANSFORM,
'exp_param': suggestion.assignments['exp_param'],
}
tunable_param_lists.append([
moving_averages,
transform_params,
suggestion.assignments['n_estimators'],
suggestion.assignments['min_samples_split'],
suggestion.assignments['min_samples_leaf'],
suggestion.assignments['bet_threshold'],
])
all_stats = read_data.generate_all_stats(box_scores)
winnings_list = runner(
historical_games_training_set,
historical_games_by_tuple,
bet_info,
all_stats,
tunable_param_lists,
)
for i, suggestion_id in enumerate(suggestion_ids):
conn.experiments(experiment.id).observations().create(
suggestion=suggestion_id,
value=winnings_list[i][0],
value_stddev=winnings_list[i][1],
)
print "Optimization done. View results at https://sigopt.com/experiment/{0}".format(
experiment.id)
experiment_detail = conn.experiments(experiment.id).fetch()
best_observation = experiment_detail.progress.best_observation
if best_observation:
print "Best value found: {0} at {1}".format(
best_observation.value, best_observation.assignments)
return experiment.id
