Defog SQLCoder Open-Source Model - Free Deployment, Instantly Convert Natural Language to High-Quality SQL Queries

Sqlcoder

Developed by defog

Defog SQLCoder is the state-of-the-art large language model for converting natural language to SQL queries, outperforming GPT-3.5-turbo and other mainstream open-source models.

Large Language Model

Transformers

Supports Multiple LanguagesOpen Source License:Other #Natural Language to SQL #Database Query Optimization #Open-source Large Models

Downloads 1,330

Release Time : 8/11/2023

Model Overview

A 15-billion-parameter model specifically designed for converting natural language to SQL queries, excelling across various database schemas.

Model Features

High-performance SQL Generation

Outperforms GPT-3.5-turbo and text-davinci-003 in the sql-eval evaluation framework

Efficient Inference

Consumes fewer GPU resources compared to models 10 times larger

Multi-database Support

Trained on 10 different database schemas with strong generalization capabilities

Phased Training

Optimized in two phases for different difficulty levels to enhance complex query capabilities

Model Capabilities

Natural Language to SQL

Complex Query Generation

Multi-table Join Processing

Grouping and Sorting Queries

Ratio Calculation

Use Cases

Data Analysis

Business Intelligence Queries

Automatically converts non-technical natural language questions into SQL queries

Reduces workload for data analysts and improves query efficiency

Database Exploration

Helps users quickly explore and understand database structures

Accelerates the learning and usage process of databases

🚀 Defog SQLCoder

Defog's SQLCoder is a cutting - edge large language model (LLM) designed to convert natural language questions into SQL queries, offering high - performance solutions for data processing.

🚀 Quick Start

This repository is now significantly outdated. You should use the repository at sqlcoder - 7b - 2 instead. It is significantly better and consumes fewer GPU resources.

Interactive Demo | ♾️ Colab | 🐦 Twitter

✨ Features

SQLCoder is a 15B parameter model that slightly outperforms gpt - 3.5 - turbo for natural language to SQL generation tasks on our sql - eval framework, and significantly outperforms all popular open - source models. It also significantly outperforms text - davinci - 003, a model that's more than 10 times its size.
SQLCoder is fine - tuned on a base StarCoder model.

Results on novel datasets not seen in training

Model	Perc Correct
gpt - 4	74.3
defog - sqlcoder	64.6
gpt - 3.5 - turbo	60.6
defog - easysql	57.1
text - davinci - 003	54.3
wizardcoder	52.0
starcoder	45.1

Results by question category

We classified each generated question into one of 5 categories. The table displays the percentage of questions answered correctly by each model, broken down by category.

Query Category	gpt - 4	defog - sqlcoder	gpt - 3.5 - turbo	defog - easy	text - davinci - 003	wizard - coder	star - coder
group_by	82.9	77.1	71.4	62.9	62.9	68.6	54.3
order_by	71.4	65.7	60.0	68.6	60.0	54.3	57.1
ratio	62.9	57.1	48.6	40.0	37.1	22.9	17.1
table_join	74.3	57.1	60.0	54.3	51.4	54.3	51.4
where	80.0	65.7	62.9	60.0	60.0	60.0	45.7

📦 Installation

You can use SQLCoder via the transformers library by downloading our model weights from the HuggingFace repo. We have added sample code for inference here. You can also use a demo on our website here, or run SQLCoder in Colab here

🔧 Technical Details

Training

Defog was trained on 10,537 human - curated questions across 2 epochs. These questions were based on 10 different schemas. None of the schemas in the training data were included in our evaluation framework.

Training happened in 2 phases. The first phase was on questions that were classified as "easy" or "medium" difficulty, and the second phase was on questions that were classified as "hard" or "extra hard" difficulty.

The results of training on our easy + medium data were stored in a model called defog - easy. We found that the additional training on hard + extra - hard data led to a 7 percentage point increase in performance.

Hardware Requirements

SQLCoder has been tested on an A100 40GB GPU with bfloat16 weights. You can also load an 8 - bit quantized version of the model on consumer GPUs with 20GB or more of memory – like RTX 4090, RTX 3090, and Apple M2 Pro, M2 Max, or M2 Ultra Chips with 20GB or more of memory.

📄 License

The model weights have a CC BY - SA 4.0 license, with OpenRAIL - M clauses for responsible use attached. The TL;DR is that you can use and modify the model for any purpose – including commercial use. However, if you modify the weights (for example, by fine - tuning), you must open - source your modified weights under the same CC BY - SA 4.0 license terms.

📋 Todo

[x] Open - source the v1 model weights
[ ] Train the model on more data, with higher data variance
[ ] Tune the model further with Reward Modelling and RLHF
[ ] Pretrain a model from scratch that specializes in SQL analysis

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご