explosion / sense2vec

🦆 Contextually-keyed word vectors
https://explosion.ai/blog/sense2vec-reloaded
MIT License
1.62k stars 240 forks source link

most_similar #120

Open mimansajaiswal opened 4 years ago

mimansajaiswal commented 4 years ago

Is there a way to specify the POS tag of the results you are looking for. Example: s2v.most_similar("dog|NOUN", n=10) returns [('puppy|NOUN', 0.9188), ('cat|NOUN', 0.9176), ('kitten|NOUN', 0.9081), ('pup|NOUN', 0.9074), ('own_dog|NOUN', 0.9028), ('other_dog|NOUN', 0.902), ('little_dog|NOUN', 0.8998), ('ferret|NOUN', 0.8952), ('chihuahua|NOUN', 0.8862), ('husky|NOUN', 0.886)] but what if I want words that are similar but are verbs. In that case I would expect to have 'chase|VERB' or 'bark|verb'. Is it possible to do that?