README.md

metadata

license: other
task_categories:
  - text-generation
  - zero-shot-classification
  - text2text-generation
language:
  - gu
tags:
  - medical
  - chemistry
  - biology
  - finance
  - legal
  - music
  - art
  - code
  - climate
pretty_name: Gujarati grammarly
size_categories:
  - 100K<n<1M

Description

This is an artificially generated list of correct-incorrect sentence pairs for Gujarati. It can be used for sentence or spelling corrections.

Use Case

1. Query correction
2. Prompt correction for language model
3. Zero-shot correction for better translation
4. Zero-shot applications to achieve similar applications for any language while training it on this dataset, as long as your model supports both languages. For example: Training an XLM-Rshared encoder-decoder model or prompt tuning a language model on this dataset and achieving inference in Italian or any other supported language, hence eliminating the need to have or generate such a dataset for the target language.