Closed demdecuong closed 4 years ago
Hi demdecuong,
The 4-gram language model is too large to keep in my personal Google Drive, so I pulled it out 3 months after my submission.
Please consider building one yourself, using the CSV V2 data from this repo https://github.com/binhvq/news-corpus, and follow the instructions from kenlm's website https://kheafield.com/code/kenlm/estimation/
Thank you for your understanding.
https://drive.google.com/file/d/14RmQSYgijeSVzZNZ2mPGL0lCLg_guXGE/view?usp=sharing