parser.add_option("-s", "--seq_size", action="store",dest="seq_size",default=False,help=u"序列长度")
(options, args) = parser.parse_args()
if options.train_file == False:
parser.print_help()
sys.exit()
if options.seq_size==False:
parser.print_help()
sys.exit()
if options.model_file==False:
parser.print_help()
sys.exit()
trainFile = options.train_file
modelFile = options.save_file
maxSeqLen = int(options.seq_size)
text = open(trainFile).read().decode("utf8")
util = TaggingUtil()
(X,Y) = util.textToTrainData(text,maxSeqLen)
taggingModel = TaggingModel()
taggingModel.loadModel(modelFile)
while True:
query = sys.stdin.readline().decode("utf8").strip()
if query == "":
break
(qX,qY) = util.textToTrainData(" ".join([w for w in query]),maxSeqLen)
qY = taggingModel.predict(qX)
result = util.trainDataToText((qX,qY), maxSeqLen)
print result
parser.print_help()
sys.exit()
if options.iter_count==False:
parser.print_help()
sys.exit()
if options.model_file==False:
parser.print_help()
sys.exit()
trainFile = options.train_file
modelFile = options.model_file
maxSeqLen = int(options.seq_size)
iterCount = int(options.iter_count)
text = open(trainFile).read().decode("utf8")
util = TaggingUtil()
(X,Y) = util.textToTrainData(text,maxSeqLen)
trainX, testX, trainY, testY = train_test_split(X, Y , train_size=0.8, random_state=1)
taggingModel = TaggingModel()
taggingModel.init(len(util.words), len(util.tags), maxSeqLen)
print "----------数据信息---------"
print " Tag数 ：",len(util.tags)
print " Word数 ：",len(util.words)
print " 训练样本数 ：",len(X)
print
print "----------数据抽样---------"
print "X ："
print X[:3]