Python NlpDict示例

示例#1

文件： mlp_se_test.py 项目： guzhuangyelao/lm-project

# -*- coding: utf-8 -*-
'''
Created on 2013-04-28 15:53
@summary: test case for MlpBigram
@author: egg
'''

from nlpdict.NlpDict import NlpDict
from pylm.MlpBigram import MlpBigram
import numpy
import time
import theano.sandbox.cuda

nlpdict = NlpDict()
nlpdict.buildfromfile('./data/pku_train_nw.ltxt')


#############
# Trainging #
#############
# text
f = file('./data/pku_train_nw.ltxt')
text = unicode(f.read(), 'utf-8')
text = text.replace(" ", "")
f.close()

len_text = len(text)

print "Train size is: %s" % len_text

theano.sandbox.cuda.use('gpu0')

示例#2

文件： rnn_se_test.py 项目： guzhuangyelao/lm-project

# -*- coding: utf-8 -*-
'''
Created on 2013-05-05 23:00
@summary: Test case on RnnLM
@author: Playcoin
'''

from nlpdict.NlpDict import NlpDict
from pylm.RnnLM import RnnLM
import numpy
import time
import theano.sandbox.cuda

theano.sandbox.cuda.use('gpu0')

nlpdict = NlpDict()
nlpdict.buildfromfile('./data/text.txt')

#############
# Trainging #
#############
# text
f = file('./data/text.txt')
text = unicode(f.read(), 'utf-8')
text = text.replace(" ", "")
f.close()

len_text = len(text)

print "Train size is: %s" % len_text

示例#3

文件： test.py 项目： guzhuangyelao/lm-project

# -*- coding: utf-8 -*-
'''
Created on 2013-04-12 23:01
@summary: 测试用例
@author: Playcoin
'''

from nlpdict.NlpDict import NlpDict
from pylm.Ngram import Ngram
import numpy

nlpdict = NlpDict()
nlpdict.buildfromfile('./data/pku_train_nw.ltxt')

# text
f = file('./data/pku_train_nw.ltxt')
text = unicode(f.read(), 'utf-8')
text = text.replace(" ", "")
f.close()

len_text = len(text)

print "Train size is: %s" % len_text

# ngram_file_path = "./data/ngram.model.obj"
ngram = Ngram(nlpdict, 2)
ngram.traintext(text)

# print "Save N-gram model"
# 
# ngram.savemodel(ngram_file_path)

示例#4

文件： rnn_lap_test.py 项目： guzhuangyelao/lm-project

# -*- coding: utf-8 -*-
'''
Created on 2013-05-05 23:00
@summary: Test case on RnnLM
@author: Playcoin
'''

from nlpdict.NlpDict import NlpDict
from pylm.RnnLM import RnnLM
import numpy
import time
import theano.sandbox.cuda

nlpdict = NlpDict()
nlpdict.buildfromfile('./data/text.txt')

#############
# Trainging #
#############
# text
f = file('./data/text.txt')
text = unicode(f.read(), 'utf-8')
text = text.replace(" ", "")
f.close()

len_text = len(text)

print "Train size is: %s" % len_text

rnnlm = RnnLM(nlpdict, n_hidden=30, lr=0.09, batch_size=50)