def train():
fileName = "trainData.model"
test = ReadFile.QAData("training.data")
test.readFile()
ProD.wordSeg(test)
ProD.delHighFre_useless(test)
ProD.delHighFre_psg(test)
trainModel(test, fileName)
def train():
fileName = "trainData.model"
test = ReadFile.QAData("training.data")
test.readFile()
ProD.wordSeg(test)
ProD.delHighFre_useless(test)
ProD.delHighFre_psg(test)
trainModel(test, fileName)
if __name__ == "__main__":
# '''
fileName = "trainData.model"
test = ReadFile.QAData("training.data")
test.readFile()
ProD.wordSeg(test)
ProD.delHighFre_useless(test)
ProD.delHighFre_psg(test)
test.calFre()
trainModel(test, fileName)
# '''
# '''
# print(model.most_similar("性格"))
# writeScore1(test, "score.data", "trainData.model")
# writeScore2(test,"score.data")
# '''
corrAns = aIndex
break
highscore = float(scoreDic[qIndex][corrAns])
flag = 1
for i in range(1, len(scoreDic[qIndex])):
if highscore < float(scoreDic[qIndex][i]):
flag = 0
showNum += 1
print("\n\n=========================================")
print(qAData.questionList[qIndex])
print(qAData.qAnswersDic[qIndex][corrAns], "\t",
scoreDic[qIndex][corrAns])
print(qAData.qAnswersDic[qIndex][i], "\t", scoreDic[qIndex][i])
break
'''
if flag==0:
print("\n\n=========================================")
print(qAData.questionList[qIndex])
for aIndex in range(len(scoreDic[qIndex])):
print(qAData.qAnswersDic[qIndex][aIndex], "\t", scoreDic[qIndex][aIndex])
'''
if showNum >= Num:
break
if __name__ == "__main__":
test = ReadFile.QAData("develop.data")
test.readFile()
scoreDic = readScores(test, "score.data")
showErr(test, scoreDic, 100)
python QA_main.py fileName scoreFile
參數說明：
fileName：問答句測試集文件名
scoreFile：得分輸出文件名
'''
import ReadFile
import ProcessData as ProD
import WordVec
import QueTypeWay
import sys
if __name__ == "__main__":
assert sys.argv.__len__() == 3
fileName = sys.argv[1]
scoreFile = sys.argv[2]
WordVec.train()
test = ReadFile.QAData(fileName)
test.readFile()
ProD.wordSeg(test)
ProD.delHighFre_useless(test)
ProD.delHighFre_psg(test)
QueTypeWay.digitToSeg(test)
test.calFre()
WordVec.writeScore1(test, scoreFile, "trainData.model")
#WordVec.writeScore2(test, scoreFile)
#WordVec.writeScore3(test, scoreFile, "trainData.model")
print("OK, DONE.")
def calQueSegFre(qAData, fileName):
freDic = {}
sum = 0
for qIndex in range(qAData.questionList.__len__()):
for tmp in qAData.questionList[qIndex]:
sum += 1
if tmp not in freDic.keys():
freDic[tmp] = 1
else:
freDic[tmp] += 1
delFile(fileName)
for key, item in sorted(freDic.items(),
key=lambda Item: Item[1],
reverse=True):
if (float(item) / freDic.__len__()) > (freThres / 10):
writeFile(fileName, [key, item, float(item) / sum])
if __name__ == "__main__":
dataFile = "training.data"
test = ReadFile.QAData(dataFile)
test.readFile()
calWordFre(test, "calWordFre.data")
ProD.wordSeg(test)
calSegFre(test, "calSegFre.data")
ProD.delHighFre_useless(test)
calQueSegFre(test, "queSegFre.data")
statQueAns(test, "statQueAns.data")
print(test.questionList.__len__())
