import codecs
import numpy as np
from sklearn.preprocessing import normalize
from generate import GENERATE
import random


vocab = codecs.open("brown_vocab_100.txt", "r", encoding="utf-16")

#load the indices dictionary
word_index_dict = {}
for i, line in enumerate(vocab):
    #TODO: import part 1 code to build dictionary

f = codecs.open("brown_100.txt", encoding = "utf-16")


counts = #TODO: initialize numpy 0s array


#TODO: iterate through file and update counts

#TODO: normalize counts


#TODO: writeout bigram probabilities



f.close()