November 2018
Beginner to intermediate
182 pages
4h 48m
English
Neither GloVe nor word2vec can handle words they didn't see during training. These words are called Out of Vocabulary (OOV), in the literature.
Evidence of this can be seen if you try to look up nouns that are not frequently used, for example an uncommon name. As you can see in the following snippet, the model throws a not in vocabulary error:
try: pretrained_w2v_model.wv.most_similar('nirant')except Exception as e: print(e) This results in the following output:
This result is also accompanied by an API warning that sometimes states the API will change in gensim v4.0.0.
Read now
Unlock full access