http://www.glozman.com/textpages.html
Tested on Python 3.5
fasttext_sg.most_similar("wizard") # FastText with Skip-Gram
>>> [('lizard', 0.9143307209014893),
('wizardkind', 0.9103794097900391),
('wiz', 0.9028722047805786),
('wizardry', 0.8854329586029053),
('triwizard', 0.8668864369392395),
('dark_wizard', 0.8515946865081787),
('wizard_chess', 0.8342143893241882),
('greatest_wizard', 0.8299335241317749),
('balding_wizard', 0.8259945511817932),
('wizard_prison', 0.8133100271224976)]
w2v_sg.most_similar("wizard") # Word2Vec with Skip-Gram
>>> [('witch', 0.7799991369247437),
('man', 0.7468042969703674),
('house_elf', 0.7278867959976196),
('witch_or', 0.7212192416191101),
('boy', 0.6939235925674438),
('lady', 0.6722564101219177),
('skinny', 0.6555918455123901),
('child', 0.652462363243103),
('thief', 0.6514541506767273),
('muggle_born', 0.6514096260070801)]
t-SNE used for dimensionality reduction
https://github.com/llSourcell/word_vectors_game_of_thrones-LIVE
http://mccormickml.com/2016/04/19/word2vec-tutorial-the-skip-gram-model/