griffith-bigdata/Qwen-2.5-Coder-0.5B-SQL-Writer
The griffith-bigdata/Qwen-2.5-Coder-0.5B-SQL-Writer is a 0.5 billion parameter language model, fine-tuned from Qwen2.5-Coder-0.5B-Instruct. This model specializes in SQL generation, having been trained on a specific text-to-SQL dataset. It is optimized for tasks requiring the conversion of natural language queries into SQL code, leveraging its compact size for efficient deployment.
Loading preview...
Model Overview
The griffith-bigdata/Qwen-2.5-Coder-0.5B-SQL-Writer is a specialized language model with 0.5 billion parameters, derived from the Qwen2.5-Coder-0.5B-Instruct architecture. Its primary function is to generate SQL queries from natural language inputs.
Key Capabilities
- SQL Generation: Fine-tuned specifically on a
sft_text2sql_v2dataset, indicating a strong focus on text-to-SQL conversion tasks. - Compact Size: With 0.5 billion parameters, it offers a relatively small footprint, suitable for applications where computational resources are a consideration.
Training Details
The model was trained using the following hyperparameters:
- Learning Rate: 2e-05
- Batch Size: 1 (train), 1 (eval)
- Gradient Accumulation: 64 steps, resulting in a total effective batch size of 128.
- Optimizer: Adam with default betas and epsilon.
- Scheduler: Cosine learning rate scheduler.
- Epochs: 2
This fine-tuning process utilized Transformers 4.45.0, Pytorch 2.2.2+cu121, Datasets 4.1.1, and Tokenizers 0.20.3.