def visualize_words(self, words_fname, palette="Viridis256"):
words = set()
for line in open(words_fname, 'r'):
if not line.startswith("#"):
for word in line.strip().split(" "):
words.add(word)
vecs = np.array([self.get_sentence_vector(word) for word in words])
visualize_words(words, vecs, palette)
def visualize_movies(self,
n_sample=30,
palette="Viridis256",
type="between"):
movie_ids = self.get_titles_in_corpus(n_sample=n_sample)
movie_titles = [movie_ids[key] for key in movie_ids.keys()]
movie_vecs = [
self.model.docvecs[self.doc2idx[movie_id]]
for movie_id in movie_ids.keys()
]
if type == "between":
visualize_between_words(movie_titles, movie_vecs, palette)
else:
visualize_words(movie_titles, movie_vecs, palette)
def my_visualize_words(self, words_set, palette="Viridis256"):
words = words_set
vecs = np.array([self.get_sentence_vector(word) for word in words])
visualize_words(words, vecs, palette, "./words.png")