

3·
2 years agoIf you’re working with a well known language, then you can probably use NLTK to tokenize your words. Word2vec is also helpful if you want a word embedding approach. https://github.com/nltk/nltk
If you’re working with a well known language, then you can probably use NLTK to tokenize your words. Word2vec is also helpful if you want a word embedding approach. https://github.com/nltk/nltk
Try looking into OpenNMT, I used it for a similar task.
I’m really encouraged by the growing number of new contributors with each post.