Hi, I came across your article on hackernoon.com (Chars2vec: character-based language model for handling real world texts with spelling errors and…). It is very interesting.
I am wondering if I can train the chars2vec model without the target (unsupervised model). Basically, I have a list of names and want to vectorize it. I have been using TF-IDF from sklearn to vectorize these names. Then, I would do some plotting, clustering, cosine similarity on these vector.
Hi, I came across your article on hackernoon.com (Chars2vec: character-based language model for handling real world texts with spelling errors and…). It is very interesting.
I am wondering if I can train the chars2vec model without the target (unsupervised model). Basically, I have a list of names and want to vectorize it. I have been using TF-IDF from sklearn to vectorize these names. Then, I would do some plotting, clustering, cosine similarity on these vector.
Thanks.