nist_nlp01_1.py

#!/usr/bin/env python3
# -*- coding: utf-8 -*-

import sys
from collections import defaultdict

def train_unigram(file_name):
  
  my_dict = defaultdict(int)

  with open(file_name, 'r') as file_open:

    for line in file_open:
      words = line.replace("\n", " \n").split(" ")
      for word in words:
        my_dict[word] += 1

  number_of_words = sum(my_dict.values())

  unigram = defaultdict(lambda: 0)

  for key in my_dict.keys():
    unigram[key] += my_dict[key]*1.0 / number_of_words

  return unigram

if __name__ == '__main__':
  unigram = train_unigram(sys.argv[1])
  print(unigram)