imahwashere/TexttoSQLish_Code

Text-to-SQL Model

A fine-tuned Llama 3.2 3B model specialized for converting natural language queries into SQL statements. This model transforms everyday questions into precise database queries, making data accessible to everyone.

Desired Use Cases

Perfect for applications requiring natural language to SQL conversion:

Business Intelligence: "Show me sales by region for last quarter" Data Analytics: "Find customers who haven't purchased in 6 months" Reporting: "What are the top 10 products by revenue?" Database Querying: Making databases accessible to non-technical users

Training Data

Trained on the gretelai/synthetic_text_to_sql dataset, which provides:

High-quality synthetic text-to-SQL pairs Diverse query patterns and complexity levels Multiple database schemas and domains Robust coverage of SQL operations and functions

Model Performance

IDK MAN, it does what it does brev. -UPDATE- YEA I didnt cook. I reckon it underfit during training/. I may have to re-do the data_prep then training and see what results the model gives us during inference and try to evaluate the model.

Uploaded model

• Developed by: imahwashere
• License: apache-2.0
• Finetuned from model : unsloth/llama-3.2-3b-instruct-unsloth-bnb-4bit

This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.