if __name__ == "__main__":
from gdayf.core.controller import Controller
from gdayf.common.constants import *
from pandas import set_option
from gdayf.common.dataload import DataLoad
#Analysis
controller = Controller()
if controller.config_checks():
data_train, data_test = DataLoad().dm()
status, recomendations = controller.exec_analysis(
datapath=data_train,
objective_column='Weather_Temperature',
amode=POC,
metric='test_rmse',
deep_impact=5)
controller.reconstruct_execution_tree(arlist=None,
metric='test_rmse',
store=True)
controller.remove_models(recomendations, mode=EACH_BEST)
set_option('display.max_rows', 500)
set_option('display.max_columns', 50)
set_option('display.max_colwidth', 100)
set_option('display.precision', 4)
set_option('display.width', 1024)
#Prediction
print('Starting Prediction\'s Phase')
if __name__ == "__main__":
from gdayf.core.controller import Controller
from gdayf.common.constants import *
from pandas import set_option
from gdayf.common.dataload import DataLoad
#Analysis
controller = Controller()
if controller.config_checks():
data_train, data_test = DataLoad().footset()
status, recomendations = controller.exec_analysis(
datapath=data_train,
objective_column='HomeWin',
amode=FAST_PARANOIAC,
metric='combined_accuracy',
deep_impact=3)
controller.reconstruct_execution_tree(metric='test_accuracy',
store=True)
controller.remove_models(arlist=recomendations, mode=EACH_BEST)
set_option('display.max_rows', 500)
set_option('display.max_columns', 50)
set_option('display.max_colwidth', 100)
set_option('display.precision', 4)
set_option('display.width', 1024)
#Prediction
print('Starting Prediction\'s Phase')
print(recomendations[0]['load_path'][0]['value'])
if __name__ == "__main__":
from gdayf.core.controller import Controller
from gdayf.common.constants import *
from pandas import set_option
from gdayf.common.dataload import DataLoad
#Analysis
controller = Controller()
if controller.config_checks():
data_train, data_test = DataLoad().dm()
status, recomendations = controller.exec_analysis(
datapath=data_train,
objective_column=None,
amode=CLUSTERING,
metric='cdistance',
deep_impact=4,
k=8,
estimate_k=False)
controller.reconstruct_execution_tree(recomendations,
metric='cdistance')
controller.remove_models(recomendations, mode=EACH_BEST)
set_option('display.max_rows', 500)
set_option('display.max_columns', 50)
set_option('display.max_colwidth', 100)
set_option('display.precision', 4)
set_option('display.width', 1024)
#Prediction
if __name__ == "__main__":
from gdayf.core.controller import Controller
from gdayf.common.constants import *
from pandas import set_option
from gdayf.common.dataload import DataLoad
#Analysis
controller = Controller()
if controller.config_checks():
data_train, data_test = DataLoad().dm()
status, recomendations = controller.exec_analysis(
datapath=data_train,
objective_column=None,
amode=ANOMALIES,
metric='train_rmse',
deep_impact=5)
controller.reconstruct_execution_tree(recomendations,
metric='train-rmse')
controller.remove_models(recomendations, mode=EACH_BEST)
set_option('display.max_rows', 500)
set_option('display.max_columns', 50)
set_option('display.max_colwidth', 100)
set_option('display.precision', 4)
set_option('display.width', 1024)
#Prediction
print('Starting Prediction\'s Phase')
"Strong-braking100", "Strong-Acel100", "idling100", "Out-speed",
"Out-engine", "Rollout100", "SDS", "SDS-slopes",
"SDS-Anticipation", "SDS-brakes", "SDS-Gear", "Fuel",
"Fuel-idling", "idling-time", "Distance-remol", "Cruise-Control",
"Avg-speed", "Max-speed", "Max-rpm", "Rollout", "Braking",
"Strong-braking", "out-rpm", "Strong-Acel100-B",
"Strong-braking100-B", "out-rpm-B", "Braking100-N", "Rollout100-N",
"Idling100-N", "Cruise-Control-N", "V_Cluster"
]
model_columns = modification(columns, ignore_columns)
print(model_columns)
status, recomendations = controller.exec_analysis(
datapath=model_data[model_columns],
objective_column=objective_column,
amode=NORMAL,
metric='test_rmse',
deep_impact=8)
controller.reconstruct_execution_tree(metric='test_rmse', store=True)
controller.remove_models(arlist=recomendations, mode=BEST)
print(
controller.table_model_list(ar_list=recomendations,
metric='test_rmse'))
prediction_frame = controller.exec_prediction(
datapath=model_data,
model_file=recomendations[0]['json_path'][0]['value'])
model_data['predict'] = prediction_frame['predict']
source_3_data = list()
if __name__ == "__main__":
from gdayf.core.controller import Controller
from gdayf.common.constants import *
from pandas import set_option
from gdayf.common.dataload import DataLoad
# Analysis
controller = Controller()
if controller.config_checks():
data_train, data_test = DataLoad().arm()
status, recomendations = controller.exec_analysis(datapath=data_train,
objective_column='ATYPE',
amode=FAST, metric='test_accuracy', deep_impact=3)
controller.reconstruct_execution_tree(metric='test_accuracy', store=True)
controller.remove_models(arlist=recomendations, mode=EACH_BEST)
set_option('display.max_rows', 500)
set_option('display.max_columns', 50)
set_option('display.max_colwidth', 100)
set_option('display.precision', 4)
set_option('display.width', 1024)
# Prediccion
print('Starting Prediction\'s Phase')
prediction_frame = controller.exec_prediction(datapath=data_test,
model_file=recomendations[0]['json_path'][0]['value'])
if 'predict' in prediction_frame.columns.values:
print(prediction_frame[['ATYPE', 'predict']])