def dictogram_dictlist(self):
""" Converts the list in word_dict to dictionary
by calling Dictogram on the values.
"""
for key, value in self.word_dict.items():
# if self.word_dict.get(key) is not None:
# print(self.word_dict.get(key))
self.word_dict[key] = dictogram.Dictogram(value)
def gen_markov(words, order=1):
pairs = make_pairs(words.split(" "), order)
markov_dict = {}
for word1, word2 in pairs:
if word1 in markov_dict:
markov_dict[word1].add_count(word2)
else:
markov_dict[word1] = dictogram.Dictogram([word2])
return markov_dict
def m_chain_one(text_list):
markov_dict = {}
for index in range(len(text_list) - 1):
window = text_list[index]
if index in markov_dict:
markov_dict[window].add_count([text_list[index + 1]])
else:
markov_dict[window] = dictogram.Dictogram([text_list[index + 1]])
return markov_dict
def order_mchain(order, text_list):
markov_dict = {}
for index in range(len(text_list) - order):
window = tuple(text_list[index: index + order])
if window in markov_dict:
markov_dict[window].add_count([text_list[index + order]])
else:
markov_dict[window] = dictogram.Dictogram([text_list[index + order]])
return markov_dict
def markov_path(token_list, order):
markov_map = {}
starting_items = []
for n in range(order):
starting_items.append(token_list[n])
state_tracker = queue.Queue(size=order, items=starting_items)
state = tuple(state_tracker.items())
markov_map[state] = dictogram.Dictogram()
for index in range(order, len(token_list)):
new_token = token_list[index]
state_tracker.enqueue(new_token)
new_state = tuple(state_tracker.items())
if new_state not in markov_map:
markov_map[new_state] = dictogram.Dictogram()
markov_map[state].add_count(new_token)
state = new_state
return markov_map
def m_chain_one(text_list):
markov_dict = {}
# for each word in list, key is word and value is dictogram
for index in range(len(text_list) - 1):
window = text_list[index]
if index in markov_dict:
markov_dict[window].add_count([text_list[index + 1]])
else:
markov_dict[window] = dictogram.Dictogram([text_list[index + 1]])
# return dictionary
return markov_dict
def new_function(self, words, order):
for i in range(len(words) - order):
if tuple(words[i:i + order]) not in self.keys():
self[tuple(words[i:i + order])] = []
self.types += 1
self[tuple(words[i:i + order])].append(
tuple(words[i + 1:i + order + 1]))
self.tokens += 1
for key in self.keys():
self[key] = dictogram.Dictogram(self[key])
def markov(list_of_words):
"""
Input: A list of tokens
Return: A sentence made by the list
provided using first order
Markov chain.
"""
dict_word = {}
word_list = []
words = list_of_words
#for each key, create a list holiding
# values that comes after it on the original list
for i in range(len(words) - 1):
if words[i] not in dict_word.keys():
word_list.append(words[i + 1])
dict_word[words[i]] = word_list
else:
dict_word[words[i]].append(words[i + 1])
word_list = []
# print(dict_word)
#make a nested dictionary using dictogram class
nested_dict = {}
for key, value in dict_word.items():
dict_hist = dictogram.Dictogram(dict_word.get(key))
nested_dict[key] = dict_hist
# print(nested_dict)
#generate sentence
sentence_list = []
first_word = random.choice(list(nested_dict.keys()))
# print("first word: "+first_word)
sentence_list.append(first_word)
# print("first word inside: " + str(nested_dict[first_word]))
dict_inside = nested_dict[first_word]
# print('weighed sample: ' + dict_inside.sample())
count = 10
while count > 0:
dict_inside = nested_dict[first_word]
next_word = dict_inside.sample()
sentence_list.append(next_word)
first_word = next_word
count -= 1
# print('Sentence list: ' + str(sentence_list))
# print('Markov Sentence: ' + ' '.join(sentence_list)+'.')
# print((' '.join(sentence_list)+'.').capitalize())
return (' '.join(sentence_list) + '.').capitalize()
def order_mchain(order, text_list):
markov_dict = {}
# for each word in list, key is word and value is dictogram
for index in range(len(text_list) - order):
# text_list[index] should be our word from list
window = tuple(text_list[index: index + order])
# check if window is stored in the dictionary already
if window in markov_dict:
# if it is, then append it to the existing histogram
markov_dict[window].add_count([text_list[index + order]])
else:
# if not, create new entry with window as key and dictogram as value
markov_dict[window] = dictogram.Dictogram([text_list[index + order]])
# return dictionary
return markov_dict
def dictogram_dictlist(self):
""" Converts the list in word_dict to dictionary
by calling Dictogram on the values.
"""
for key, value in self.word_dict.items():
self.word_dict[key] = dictogram.Dictogram(value)
def generate(token_list):
histogram = dictogram.Dictogram(token_list)
return histogram