Closed fanweihua closed 3 years ago
Hi @fanweihua,
Thanks for your issue! Unfortunately, we don't have such a feature. Could you please elaborate a bit more about your usecase?
I've been evaluating deepcut and attacut. I was impressed by attacut's speed. But I need the feature for custom dictionary if the tokenization result is not what I expect.
this repo might be useful for you in this case; you can use it to post-process the tokenized text.
https://github.com/heytitle/mudyom
Feel free ping me if you have any further question!
Is there a way to add custom dictionary? like deepcut?