3 ms·

Is this the data that was used for fine tuning? https://github.com/defog-ai/sql-eval/blob/main/data/questions_gen_postgres.csv https://github.com/defog-ai/sql-

• thecalebf 3y ago
  No, those are benchmark, evaluation questions. The fine tune dataset was a custom, synthetically generated dataset of ~20k PostgreSQL Text to SQL pairs covering different SQL categories and question types. I mention a little more about it here https://x.com/calebfahlgren/status/1754247740291207198?s=20 https://x.com/calebfahlgren/status/1754247740291207198?s=20
  • Semaphor 3y ago
    So this is essentially postgres only? Or how will it handle e.g. MS SQL Schemas and output?
    • thecalebf 3y ago
      Currently Postgres yes, already working on a dataset with more DDLs like MySQL, DuckDB, MSSQL, etc for a second iteration.