def __init__(self,
dataID,
b_allow_repeated_value=True,
logger_file=logger_file,
num_of_points=None,
base_models=[tools.app_linear]):
self.logger = logs.QueryLogs(log=logger_file)
# self.logger.set_no_output()
self.data = dl.load2d(dataID)
if not b_allow_repeated_value:
self.data.remove_repeated_x_1d()
self.cRegression = CRegression(logger_object=self.logger,
base_models=base_models)
self.cRegression.fit(self.data)
# self.logger.set_logging(file_name=logger_file)
if num_of_points is None:
self.qe = QueryEngine(self.cRegression, logger_object=self.logger)
else:
self.qe = QueryEngine(self.cRegression,
logger_object=self.logger,
num_training_points=num_of_points)
self.qe.density_estimation()
self.q_min = min(self.data.features)
self.q_max = max(self.data.features)
self.dataID = dataID
del self.data
def __init__(self,logger_object=None):
self.num_datasetaset = 8
if not logger_object:
logger_object = logs.QueryLogs()
self.logger = logger_object
self.logger_name = logger_object.logger_name
def __init__(self, dataset="tpcds", logger_file=logger_file,base_models=[tools.app_xgboost]):
self.dataset = dataset
self.logger = logs.QueryLogs(log=logger_file)
self.logger.set_level("INFO")
self.logger.logger.info("Initialising DBEst...")
self.group_names = {}
self.num_of_points_per_group_tbls = {}
self.num_of_points = {}
self.df = {}
self.DBEstClients = {}
self.tableColumnSets = [] # store all QeuryEngines, for each table
self.base_models = base_models
def __init__(self,
cregression,
logger_object=None,
b_print_time_cost=True,
num_training_points=None):
if num_training_points is None:
self.num_training_points = cregression.num_total_training_points
else:
self.num_training_points = num_training_points
self.log_dens = None
self.training_data = cregression.training_data
self.kde = None # kernel density object
self.dimension = self.training_data.features.shape[1]
self.cregression = cregression
if logger_object:
self.logger = logger_object.logger
else:
self.logger = logs.QueryLogs().logger
self.b_print_time_cost = b_print_time_cost
warnings.filterwarnings(action='ignore', category=DeprecationWarning)
q_max_boundary,
steps=1000,
n_bisect=50)
end = datetime.now()
time_cost = (end - start).total_seconds()
if self.b_print_time_cost:
self.logger.info("Time spent for approximate CORR: %.4fs." %
time_cost)
return result, time_cost
if __name__ == "__main__":
import generate_random
warnings.filterwarnings(action='ignore', category=DeprecationWarning)
logger = logs.QueryLogs()
logger.set_no_output()
data = dl.load2d(5)
cRegression = CRegression(logger_object=logger)
cRegression.fit(data)
# cRegression.plot_training_data_2d()
logger.set_logging()
qe = QueryEngine(cRegression, logger_object=logger)
qe.density_estimation()
# qe.desngity_estimation_plt2d()
r = generate_random.percentile(0.9,
qe.kde,
30,
100,