BeachWang/DAIL-SQL

Steps to Evaluate DAIL-SQL on Custom Data

Opened this issue · 1 comments

Thank you for the great work on DAIL-SQL!

I’m interested in using DAIL-SQL on some custom data that I have. Specifically, I have a dataset with natural language questions and corresponding database schemas, and I’d like to evaluate the system’s performance on generating SQL queries for my inputs.

Could you please provide detailed steps on how to:

  1. Prepare the input data in the required format?
  2. Configure the system to process custom inputs?
  3. Run the system to generate SQL queries for the input data?

Thank you in advance for your help!

Hi, Thank you for your interest in DAIL-SQL!

I think you can manage your data as the format of Spider Dev dataset. For convenience, you can also use the Spider Train data as the candidate set as we do in our experiments. Then you can just follow the README.md to generate SQL queries.

Hope these suggestions are helpful to you.