from Data.DataManager import DataManager
from ScriptToolkit import ScriptToolkit
from Preprocessing.DataReader import DataReader
from Preprocessing import ProcessorFactory
from Model.ConditionalRandomField import CRF
if __name__ == '__main__':
# create data manager
DM = DataManager()
DM.change_pwd()
DM.source_data_file = 'CorpusLabelData_SalesModule.txt'
DM.remove(DM.log_wrong_sentences)
# create datums
DR = DataReader(source_data_file=DM.source_data_file)
DR.standard_read()
# create toolkits
ST = ScriptToolkit(DM)
features = ScriptToolkit.get_demo_features()
# analysis
sent_accuracys, train_times, test_times = [], [], []
cycle_times = 30
for i in range(cycle_times):
# data preprocessing
crf_processor = ProcessorFactory.CRFProcessorFactory().produce(
source_data_file=DM.source_data_file,
train_file=DM.train_file,
test_file=DM.test_file)
crf_processor.get_train_data(DR.Datums)
from Data.DataManager import DataManager
from Preprocessing import ProcessorFactory
from Model.ConditionalRandomField import CRF
from ScriptToolkit import ScriptToolkit
from Preprocessing.DataReader import DataReader
if __name__ == '__main__':
# create data manager
DM = DataManager()
DM.change_pwd()
DM.source_data_file = 'CorpusLabelData_MergedFilter_Update.txt'
DM.remove(DM.log_wrong_sentences)
# create datums
DR = DataReader(source_data_file=DM.source_data_file)
DR.standard_read()
# create toolkits
ST = ScriptToolkit(DM)
features = ScriptToolkit.get_demo_features()
# analysis
sent_accuracys = []
cycle_times = 1
for i in range(cycle_times):
# data preprocessing
crf_processor = ProcessorFactory.CRFProcessorFactory().produce(
source_data_file=DM.source_data_file,
train_file=DM.train_file,
test_file=DM.test_file)
crf_processor.get_train_data(DR.Datums)