This repository contains code for dataset creation/scraping for the low resource language Urdu. Furthermore, it also contains code for training and testing the model google/mt5-base on the datasets generated. The datasets that we scraped/generated are also included. We will restructure the repository and write a guide for reproducing our results soon. You can check out live demo of the model here: https://qawaid.vercel.app
mh07607/urdu_grammar_correction
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|