| --- |
| base_model: mistralai/Mistral-7B-Instruct-v0.3 |
| datasets: |
| - generator |
| library_name: peft |
| license: apache-2.0 |
| tags: |
| - trl |
| - sft |
| - generated_from_trainer |
| model-index: |
| - name: Mistral-7B-Text2SQL |
| results: [] |
| --- |
| |
| <!-- This model card has been generated automatically according to the information the Trainer had access to. You |
| should probably proofread and complete it, then remove this comment. --> |
|
|
| # Mistral-7B-Text2SQL |
|
|
| This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.3](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3) on the generator dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 0.4643 |
|
|
| ## Model description |
|
|
| This repository contains a fine-tuned version of the Mistral 7B model, tailored specifically for text-to-SQL tasks. |
| The model is designed to convert natural language queries into structured SQL queries, enabling seamless interaction with databases through conversational language. |
|
|
| ## Intended uses & limitations |
|
|
| The Mistral-7B-Text2SQL model is intended for applications that require converting natural language queries into SQL commands. Suitable use cases include: |
|
|
| Conversational Agents: Allowing users to retrieve information from databases through natural language interaction. |
| Data Analytics: Enabling non-technical users to query databases without needing to know SQL syntax. |
| Business Intelligence: Supporting decision-making processes by simplifying data access. |
|
|
| ## Training and evaluation data |
|
|
| The model was fine-tuned using the generator dataset, which consists of a variety of natural language queries paired with corresponding SQL commands. The dataset is designed to cover a wide range of query types, allowing the model to generalize better across different types of SQL queries. |
|
|
| Dataset Characteristics |
| Diversity: The dataset includes examples from various domains, ensuring that the model learns to handle a broad spectrum of queries. |
| Size: (Include the size of the dataset, e.g., the number of examples if available.) |
| Annotations: Each example includes natural language input along with the expected SQL output, facilitating supervised learning. |
|
|
|
|
|
|
| ### Training results |
|
|
| | Training Loss | Epoch | Step | Validation Loss | |
| |:-------------:|:-----:|:----:|:---------------:| |
| | 1.8346 | 0.4 | 10 | 0.7031 | |
| | 0.5882 | 0.8 | 20 | 0.5273 | |
| | 0.487 | 1.2 | 30 | 0.4850 | |
| | 0.4423 | 1.6 | 40 | 0.4675 | |
| | 0.4235 | 2.0 | 50 | 0.4564 | |
| | 0.3464 | 2.4 | 60 | 0.4690 | |
| | 0.3411 | 2.8 | 70 | 0.4643 | |
|
|
|
|
| ### Framework versions |
|
|
| - PEFT 0.13.2 |
| - Transformers 4.45.2 |
| - Pytorch 2.4.1+cu121 |
| - Datasets 3.0.1 |
| - Tokenizers 0.20.1 |
