def txtED(s):
    container = set()
    word_list = s.split()
    for i in word_list:
      a = i.lower()
      b = a.join([c for c in s if c.isalpha()])
      container.add(b)
    return container


txtFile3  = sc.wholeTextFiles(args.input)
        numOfDoc = txtFile3.keys().count()
        RDD = txtFile3.map(lambda (doc, vals):(doc, txtED(vals))).values().flatMap(lambda word: (word, 1))\
               .saveAsTextFile(args.output)