text2sql / README.md
juanfra218's picture
Update README.md
6d446a9 verified
|
raw
history blame
2.78 kB
metadata
license: mit

Fine-Tuned Google T5 Model for Text to SQL Translation

This repository contains a fine-tuned version of the Google T5 model, specifically trained for the task of translating natural language queries into SQL statements.

Model Details

Fine-Tuning Datasets

  1. sql-create-context Dataset:

    • This dataset was created by modifying data from the following sources:
      • Zhong, Victor, et al. "Seq2SQL: Generating Structured Queries from Natural Language using Reinforcement Learning." (2017).
      • Yu, Tao, et al. "Spider: A large-scale human-labeled dataset for complex and cross-domain semantic parsing and text-to-SQL task." (2018).
    • Citation:
      @misc{b-mc2_2023_sql-create-context,
        title   = {sql-create-context Dataset},
        author  = {b-mc2}, 
        year    = {2023},
        url     = {https://huggingface.co/datasets/b-mc2/sql-create-context},
        note    = {This dataset was created by modifying data from the following sources: \cite{zhongSeq2SQL2017, yu2018spider}.},
      }
      
  2. Synthetic-Text-To-SQL Dataset:

    • A synthetic dataset for training language models to generate SQL queries from natural language prompts.
    • Citation:
      @software{gretel-synthetic-text-to-sql-2024,
        author = {Meyer, Yev and Emadi, Marjan and Nathawani, Dhruv and Ramaswamy, Lipika and Boyd, Kendrick and Van Segbroeck, Maarten and Grossman, Matthew and Mlocek, Piotr and Newberry, Drew},
        title = {{Synthetic-Text-To-SQL}: A synthetic dataset for training language models to generate SQL queries from natural language prompts},
        month = {April},
        year = {2024},
        url = {https://huggingface.co/datasets/gretelai/synthetic-text-to-sql}
      }
      

Ongoing Work

I am currently working to implement PICARD (Parsing Incrementally for Constrained Auto-Regressive Decoding from Language Models) to improve the results of this model. More details can be found in the original PICARD paper:

  • Citation:
    @misc{scholak2021picardparsingincrementallyconstrained,
          title={PICARD: Parsing Incrementally for Constrained Auto-Regressive Decoding from Language Models}, 
          author={Torsten Scholak and Nathan Schucher and Dzmitry Bahdanau},
          year={2021},
          eprint={2109.05093},
          archivePrefix={arXiv},
          primaryClass={cs.CL},
          url={https://arxiv.org/abs/2109.05093}, 
    }