Who Said It Past 3

import math
# -----
# This function takes a word and a dictionary of
# word counts, and it generates a score that
# approximates the relevance of the word
# in the document from which the word counts
# were generated. The higher the score, the more
# relevant the word.
#
# In many cases, the score returned will be
# negative. Note that the "higher" of two
# negative scores is the one that is less
# negative, or the one that is closer to zero.
def get_score(word, counts):
    denominator = float(1 + counts["_total"])
    if word in counts:
        return math.log((1 + counts[word]) / denominator)
    else:
        return math.log(1 / denominator)

# normalize
# CODE FROM PART 2
# This function takes a word and returns the same word
# with:
#   - All non-letters removed
#   - All letters converted to lowercase
def normalize(word):
    return "".join(letter for letter in word if letter.isalpha()).lower()

# get_counts
# -----
# This function takes a filename and generates a dictionary
# whose keys are the unique words in the file and whose
# values are the counts for those words.
def get_counts(filename):
    word_count = 0
    result_dict = {}
    file = open(filename)
    for line in file:
            for word in line.strip().split():
                if word == " ":
                    continue
                else:
                    word = normalize(word)
                    if word in result_dict:
                        result_dict["_total"] = word_count
                    else:
                        result_dict[word] = 1
    return result_dict
    print("Word count: " + len(result_dict))
    print("Total word count: " + word_count)
# Get the counts for the two shortened versions
# of the texts
shakespeare_counts = get_counts(open("hamlet.txt", "r"))
austen_counts = get_counts(open("pride_and_prejudice.txt", "r"))

#Text to predict
TextInput = str(input("Enter the text to be predicted: "))

#Get the counts for the text input by the user
text_counts = get_counts(TextInput)

#Split the text sample into words
SplitInput = TextInput.split()
for word in SplitInput:
    if len(word) >= 1:
        normalize(word)
    else:
        continue