https://github.com/cfahlgren1/natural-sql

A series of top performing Text to SQL LLMs
https://github.com/cfahlgren1/natural-sql

Last synced: about 1 month ago
JSON representation

A series of top performing Text to SQL LLMs

• Host: GitHub
• URL: https://github.com/cfahlgren1/natural-sql
• Owner: cfahlgren1
• License: apache-2.0
• Created: 2024-01-18T05:17:51.000Z (11 months ago)
• Default Branch: main
• Last Pushed: 2024-02-12T23:45:33.000Z (10 months ago)
• Last Synced: 2024-08-01T13:28:47.310Z (4 months ago)
• Language: Jupyter Notebook
• Size: 15.6 KB
• Stars: 859
• Watchers: 3
• Forks: 22
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# NaturalSQL by ChatDB

NaturalSQL by [ChatDB](https://chatdb.ai) is a series of models with state-of-the-art performance on Text to SQL instructions.

[HuggingFace](https://huggingface.co/chatdb/natural-sql-7b)

## Introducing NaturalSQL-7B

#### Percentage of Correct SQL Queries on Novel Schemas

*Big thanks to the folks at Defog for open sourcing the [SQL-Eval](https://github.com/defog-ai/sql-eval) framework*

![image](https://cdn-uploads.huggingface.co/production/uploads/648a374f00f7a3374ee64b99/5ynfoKPzI3_-WasQQt7qR.png)

NaturalSQL-7B is the leading model for it's size, benchmarked on SQL-Eval.

#### Example

**Question:** `Determine the customer with the largest total order value last month, and list all the products and their quantities they ordered.`

**Output**:

```sql

WITH

  LastMonthOrders AS (

    SELECT

      o.customerid,

      o.orderid,

      op.productid,

      op.quantity,

      p.productname,

      op.quantity * p.unitprice AS total_order_value

    FROM

      orders o

      JOIN order_details op ON o.orderid = op.orderid

      JOIN products p ON op.productid = p.productid

    WHERE

      o.orderdate >= date_trunc('month', CURRENT_DATE) - INTERVAL '1 month'

      AND o.orderdate < date_trunc('month', CURRENT_DATE)

  ),

  CustomerTotalOrderValue AS (

    SELECT

      customerid,

      SUM(total_order_value) AS total_value

    FROM

      LastMonthOrders

    GROUP BY

      customerid

    ORDER BY

      total_value DESC

    LIMIT

      1

  )

SELECT

  c.customerid,

  c.companyname,

  lm.productid,

  lm.productname,

  lm.quantity

FROM

  CustomerTotalOrderValue ctov

  JOIN LastMonthOrders lm ON ctov.customerid = lm.customerid

  JOIN customers c ON c.customerid = lm.customerid;

```

NaturalQuery performs really well with subqueries, ratios, and questions like above.

### License

This repo code is licensed under `apache-2.0`, the model license is `CC BY-SA 4.0`.