#!/usr/bin/env python # -*- coding: utf-8 -*- #coding=gbk from PreProcesser import XmlConverter xmlConverter = XmlConverter() #xmlConverter.convertDoc(99900, 99900) xmlConverter.convertQuery(19,21)
f.write(content)
if __name__ == "__main__":
reload(sys)
sys.setdefaultencoding('utf8')
initDirectories()
handleArgv()
if Config.FEEDBACK_MODE == 1:
feedbackGenerator = FeedbackQueryGenerator()
feedbackGenerator.genFeedbackQuery()
sys.exit(0)
if Config.PRE_PROCESS_ON == 1:
xmlConverter = XmlConverter()
xmlConverter.convertDoc(0, Config.DATA_SIZE)
xmlConverter.convertQuery(0, Config.QUERY_SIZE)
else:
docReader = DocReader()
docModeler = DocModeler()
for d in range(0, Config.TEST_DATA_SIZE):
words = docReader.loadSegDoc(d)
docModeler.genModelByDocArr(words, d)
print "Done generate %d model" % (d)
queryResult = []
for q in range(1, 11):
scoreList = []
fw.write('NewsId,Agency\n')
else:
eachLine = eachLine.replace("\n", "")
content = '%s,%s\n' % (eachLine, clfResults[resultCounter])
fw.write(content)
resultCounter += 1
if __name__ == "__main__":
reload(sys)
sys.setdefaultencoding('utf8')
initDirectories()
handleArgv()
if Config.PRE_PROCESS_ON == 1:
xmlConverter = XmlConverter()
xmlConverter.convertDoc(0, Config.DATA_SIZE)
xmlConverter.convertQuery(0, Config.QUERY_SIZE)
xmlConverter.convertTestData(0, Config.TEST_DATA_SIZE)
else:
docReader = DocReader()
docModeler = DocModeler()
trainDataReader = TrainDataReader()
featureModeler = FeatureBasedModeler()
Y, trainDataIdxs = trainDataReader.getTrainAnswers()
print "Get Train Answers Done"
tfidfMat, Y = featureModeler.extractFeaturesMatrix(trainDataIdxs, Y)
print "Calc data features done"
import sys
import os
import numpy
import Config
from PreProcesser import XmlConverter
from PreProcesser import DocReader
from Modeler import DocModeler
if __name__ == "__main__":
reload(sys)
sys.setdefaultencoding('utf8')
if Config.PRE_PROCESS_ON == 1:
xmlConverter = XmlConverter()
#xmlConverter.convertDoc(0, Config.DATA_SIZE)
xmlConverter.convertQuery(0, Config.QUERY_SIZE)
else:
docReader = DocReader()
docModeler = DocModeler()
for d in range(0, Config.TEST_DATA_SIZE):
words = docReader.loadSegDoc(d)
docModeler.genModelByDocArr(words, d)
print "Done generate %d model" % (d)
scoreList = []
for q in range(5, 6):
query = docReader.loadQuery(q)
for d in range(0, Config.TEST_DATA_SIZE):