Model Information
This model, Qwen-3.5-4B-Instruct-Spatial-SQL-1.1, is an 4B, narrow use case, text to spatial SQL, lightly fine-tuned model. In general, its primary use case is the Natural Language command adaptation of particular geographic spatial functions as normally defined in pure SQL. Data input should be a combination of an English prefix in the form of a question, and a coordinate prompt injection, likely from an active mapping system application coordinate list. Output is PostGIS spatial SQL.
There are ten geographic functions released in version 1.2.
Model developer: Mark Rodrigo
Github: https://github.com/mprodrigo/spatialsql
Model Architecture: The model is a QLoRA / Supervised Fine Tuning (SFT)
Model Input / Output Overview:
Example Prompt / Prompt File
Examples
A Few Known Question Variation Examples
llama.cpp / Hyperparameter Recommendations For Inference
Agent Considerations
Further Reference - link this
Evaluation data
Training data
Custom synthetic
Training hyperparameters
Training results
| Training Loss | Step | Validation Loss |
|---|---|---|
| 1.1217 | 10 | 0.8837 |
| 0.8026 | 20 | 0.7469 |
| 0.7432 | 30 | 0.6925 |
| 0.6930 | 40 | 0.6455 |
| 0.6189 | 50 | 0.5935 |
| 0.5665 | 60 | 0.5342 |
| 0.4824 | 70 | 0.4802 |
| 0.4461 | 80 | 0.4557 |
Framework versions
- Transformers 4.46.1
- Pytorch 2.5.1
- peft 0.19.1
- Datasets 5.0.0
- Tokenizers 0.22.2