コード例 #1
0
ファイル: wordCloudDraw.py プロジェクト: BloodD/PyTopicM
 def createTagCloud(self,wordline):
     """
     Create tag cloud image 
     """
     wordstream = []
     if wordline == '':
         return False
     
     wordsTokens = WhitespaceTokenizer().tokenize(wordline)
     wordsTokens.remove(wordsTokens[0])
     wordstream.append(' '.join(wordsTokens))
     wordstream = ' '.join(wordstream)
     thresh = self.wordCount
     colorS = self.colorSchemes[self.color]
     
     tags = make_tags(get_tag_counts(wordstream)[:thresh],\
                      minsize=3, maxsize=40,\
                      colors = COLOR_SCHEMES[colorS])
     
     create_tag_image(tags, self.png,\
                      size=(960, 400),\
                      background=(255, 255, 255, 255),\
                      layout= LAYOUT_HORIZONTAL,\
                      fontname='Neuton')
     return True
コード例 #2
0
ファイル: wordpipemodel.py プロジェクト: BloodD/PyTopicM
 def topicWordFreq(self,wordline):
     """
     Statistic word stream frequency.
     """
     wordstream = []
     wordsTokens = WhitespaceTokenizer().tokenize(wordline)
     wordsTokens.remove(wordsTokens[0])
     wordstream.append('  '.join(wordsTokens))
     wordstream = ' '.join(wordstream)
     #print wordstream
     
     fdist = FreqDist()
     for word in word_tokenize(wordstream):
         fdist.inc(word)
     result = [list(item) for item in fdist.items()]
     num = float(fdist.N())
     result = [[val[0],val[1],val[1]/num] for val in result]
     #print "smaples:"
     #print fdist.items()
     #print fdist.keys()
    
     return result