def online_traning_api(input_file_name,model_id):
print '%s' %settings.logging_file_training
utils.setLog(settings.logging_file_training)
logger=logging.getLogger('model-learner.train')
#Log start of Full process
utils.logInfoTime(logger, 'Started')
# run hbc load data script
logger.info('==> Load Data.')
utils.logInfoTime(logger, 'Started Data Load')
data_np_array, y_np_array = data_load_csv.csv_train_from_one_file(input_file_name);
utils.logInfoTime(logger, 'Finished Data Load')
# preprocessing featurizer sample data
logger.info('==> Preprocessing feature data.')
utils.logInfoTime(logger, 'Started Preprocessing')
utils.logInfoTime(logger, 'Finished Preprocessing')
# build models
logger.info('==> Build Model.')
utils.logInfoTime(logger, 'Started Model Building')
model_building.modelsBuild(data_np_array, y_np_array,model_id,logger)
utils.logInfoTime(logger, 'Finished Model Building')
utils.logInfoTime(logger, 'Finished')
import datetime
import data_load
import data_preprocessing
import model_building
import utils
import pprint as pp # to make log entries nicer to read
# get the settings for the run
import settings
#Setup the logger
utils.setLog(settings.LOGGING_FILE)
logger=logging.getLogger('Master')
#Log start of Full process
utils.logInfoTime(logger, 'Started')
# run data load script
logger.info('--------------------------------- Data Load -----------------------------------')
utils.logInfoTime(logger, 'Started Data Load')
# initial_data = data_load.psqlLoad(settings.INPUT_TABLE, settings.INPUT_SCHEMA, columns='*')
initial_data = data_load.csvfile(settings.INPUT_DIR, settings.file_name, settings.RESULTS_OUTPUT_DIR)
utils.logInfoTime(logger, 'Finished Data Load')
# run preprocessing script
logger.info('--------------------------------- Data Preprocessing -----------------------------------')
utils.logInfoTime(logger, 'Started Pre-Processing')
data_np_array, y_np_array, var_results = data_preprocessing.main(initial_data)
utils.logInfoTime(logger, 'Finished Pre-Processing')
def offline_train():
#Setup the logger
print '%s' %settings.logging_file_training
utils.setLog(settings.logging_file_training)
logger=logging.getLogger('model-learner.train')
#Log start of Full process
utils.logInfoTime(logger, 'Started')
# run hbc load data script
logger.info('==> Load Data.')
utils.logInfoTime(logger, 'Started Data Load')
data_np_array, y_np_array = data_load_csv.csv_train_file(settings.INPUT_DIR, settings.train_file_name_white, settings.train_file_name_black)
utils.logInfoTime(logger, 'Finished Data Load')
# preprocessing featurizer sample data
logger.info('==> Preprocessing feature data.')
utils.logInfoTime(logger, 'Started Preprocessing')
utils.logInfoTime(logger, 'Finished Preprocessing')
# build models
logger.info('==> Build Model.')
utils.logInfoTime(logger, 'Started Model Building')
model_building.modelsBuild(data_np_array, y_np_array, 'hbc_train_offline.model',logger)
utils.logInfoTime(logger, 'Finished Model Building')
utils.logInfoTime(logger, 'Finished')
print('model training complete')
