Little snippet to create a word count dictionary

create_word_count_dictionary.py

def conta_palavras(text, stop_words, punctuations):
  #divide o texto em tokens
  tokens = word_tokenize(text)
  #remove as stopwords e pontuacao
  keywords = [word for word in tokens if not word in stop_words and not word in punctuations]

  #gera dicionario com as palavras e contagem
  wordcount = {}
  for word in keywords:
      word = word.lower() #todas minúsculas
      if word not in stop_words:
          if word not in wordcount:
              wordcount[word] = 1
          else:
              wordcount[word] += 1

  return wordcount