seed = random.randrange(n-1, size)
for i in range(seed - (n-1), seed):
context.append(wordlist[i])
return context, wordlist[seed][0]
# corpus, trainCorpus, testCorpus, testSents = corpus.loadCorpus()
corpus, trainCorpus, testCorpus, testSents = corpus.loadCorpus("POP")
# corpus, trainCorpus, testCorpus, testSents = corpus.loadCorpus("ROCK")
# print (testCorpus)
nTag = 2
nWord = 4
n = max(nTag, nWord)
tm = NgramTagModel(nTag,nWord,trainCorpus,0.5)
# testTag = tm.tagTestCorpus(testCorpus)
totalTest = 0
correctTest = 0
for sent in testSents:
tagSent = tm.tagTestCorpus(sent)
if(len(list(sent)) >= n):
print (sent)
context, correctWord = getWordContext(tagSent, n)
print (context, correctWord)
predictWord = tm.nextWord(context)
print (predictWord)
totalTest += 1
if(totalTest >= 100):
break
__author__ = 'Salma'
#from nGram import NgramTagModel,nGramModel
import corpus.lyric_corpus.corpus_access as corpus
from nGram.NgramTagModel import NgramTagModel
trainCorpus, testCorpus, devCorpus = corpus.loadCorpus("POP")
# trainCorpus, testCorpus, devCorpus = corpus.loadCorpus("ROCK")
# print (testCorpus)
tm = NgramTagModel(3,2,trainCorpus)
testTag = tm.tagTestCorpus(testCorpus)
context = tm.getRandomContext(testTag)
print("Context", context)
print(tm.nextWord(context))
# print(sents)