gpt-pydantic-tools


Namegpt-pydantic-tools JSON
Version 0.2 PyPI version JSON
download
home_pageNone
SummaryDefine GPT tool schemas using Pydantic.
upload_time2024-05-19 17:41:52
maintainerNone
docs_urlNone
authorNone
requires_pythonNone
licenseMIT License
keywords gpt ai tool schemas pydantic json
VCS
bugtrack_url
requirements No requirements were recorded.
Travis-CI No Travis.
coveralls test coverage No coveralls.
            # GPT Pydantic Tools

The `gpt_pydantic_tools` repository offers a Python module designed to integrate Pydantic models with GPT-style tool schemas. It facilitates the transformation of Pydantic models into a format that is suitable for use with GPT-4's tools functionality, ensuring structured data and tool interaction consistency.

## Features

- **Model Conversion:** Convert Pydantic models into GPT-4 tool schemas.
- **Flexible Tool Choice Handling:** Provides several strategies for tool invocation such as auto, required, and none, based on the context or specific requirements.

## Installation

Install directly using pip:
```bash
pip install gpt-pydantic-tools
```

## Usage

1. **Define Pydantic Models:** Create your Pydantic models as per your requirements.
2. **Convert to GPT Tool Schema:** Use the `ToolSchemaManager` to convert your Pydantic models into GPT tool schemas.
3. **Tool Choice Management:** Utilize the `get_tool_choice_dict` to manage how tools are chosen for execution based on the defined strategies.

Example usage:
```python
from pydantic import BaseModel
from gpt_pydantic_tools import ToolSchemaManager, get_tool_choice_dict, ToolChoiceEnum

class MyModel(BaseModel):
    name: str
    age: int

# Convert Pydantic model to GPT tool schema
schema_manager = ToolSchemaManager(pydantic_obj=MyModel)
tool_schema = schema_manager.tools_schema

# Determine tool invocation strategy
tool_choice = get_tool_choice_dict(ToolChoiceEnum.AUTO, schema_manager)
```

If you want to do the Pydantic Schema conversion yourself, and directly provide the JSON instead, you can do:
```
schema_manager = ToolSchemaManager(pydantic_obj_json_schema=my_model_json_schema)
```

## Contributing

Contributions are welcome! Please feel free to submit pull requests, create issues, and suggest improvements to the repository.

## License

This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.

            

Raw data

            {
    "_id": null,
    "home_page": null,
    "name": "gpt-pydantic-tools",
    "maintainer": null,
    "docs_url": null,
    "requires_python": null,
    "maintainer_email": null,
    "keywords": "gpt, ai, tool, schemas, pydantic, json",
    "author": null,
    "author_email": "\"Carlos A. Planch\u00f3n\" <carlosandresplanchonprestes@gmail.com>",
    "download_url": "https://files.pythonhosted.org/packages/1e/b1/bc022cc16e6f3a3b46ae59d52ff996ea63e1527375755f94641a311633a2/gpt_pydantic_tools-0.2.tar.gz",
    "platform": null,
    "description": "# GPT Pydantic Tools\n\nThe `gpt_pydantic_tools` repository offers a Python module designed to integrate Pydantic models with GPT-style tool schemas. It facilitates the transformation of Pydantic models into a format that is suitable for use with GPT-4's tools functionality, ensuring structured data and tool interaction consistency.\n\n## Features\n\n- **Model Conversion:** Convert Pydantic models into GPT-4 tool schemas.\n- **Flexible Tool Choice Handling:** Provides several strategies for tool invocation such as auto, required, and none, based on the context or specific requirements.\n\n## Installation\n\nInstall directly using pip:\n```bash\npip install gpt-pydantic-tools\n```\n\n## Usage\n\n1. **Define Pydantic Models:** Create your Pydantic models as per your requirements.\n2. **Convert to GPT Tool Schema:** Use the `ToolSchemaManager` to convert your Pydantic models into GPT tool schemas.\n3. **Tool Choice Management:** Utilize the `get_tool_choice_dict` to manage how tools are chosen for execution based on the defined strategies.\n\nExample usage:\n```python\nfrom pydantic import BaseModel\nfrom gpt_pydantic_tools import ToolSchemaManager, get_tool_choice_dict, ToolChoiceEnum\n\nclass MyModel(BaseModel):\n    name: str\n    age: int\n\n# Convert Pydantic model to GPT tool schema\nschema_manager = ToolSchemaManager(pydantic_obj=MyModel)\ntool_schema = schema_manager.tools_schema\n\n# Determine tool invocation strategy\ntool_choice = get_tool_choice_dict(ToolChoiceEnum.AUTO, schema_manager)\n```\n\nIf you want to do the Pydantic Schema conversion yourself, and directly provide the JSON instead, you can do:\n```\nschema_manager = ToolSchemaManager(pydantic_obj_json_schema=my_model_json_schema)\n```\n\n## Contributing\n\nContributions are welcome! Please feel free to submit pull requests, create issues, and suggest improvements to the repository.\n\n## License\n\nThis project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.\n",
    "bugtrack_url": null,
    "license": "MIT License",
    "summary": "Define GPT tool schemas using Pydantic.",
    "version": "0.2",
    "project_urls": {
        "repository": "https://github.com/carlosplanchon/gpt_pydantic_tools.git"
    },
    "split_keywords": [
        "gpt",
        " ai",
        " tool",
        " schemas",
        " pydantic",
        " json"
    ],
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "blake2b_256": "5a6283ecf6e158fe41bfcaf683a9de87a1654dc8ec8789578f4ed11f32296d1a",
                "md5": "be9e6b59322c4692a3929d892f7a3a00",
                "sha256": "24ca9549ee89c1694dc5526c2c7ae03cea82045b572606fee7452318b46b7ed8"
            },
            "downloads": -1,
            "filename": "gpt_pydantic_tools-0.2-py3-none-any.whl",
            "has_sig": false,
            "md5_digest": "be9e6b59322c4692a3929d892f7a3a00",
            "packagetype": "bdist_wheel",
            "python_version": "py3",
            "requires_python": null,
            "size": 4716,
            "upload_time": "2024-05-19T17:41:49",
            "upload_time_iso_8601": "2024-05-19T17:41:49.545202Z",
            "url": "https://files.pythonhosted.org/packages/5a/62/83ecf6e158fe41bfcaf683a9de87a1654dc8ec8789578f4ed11f32296d1a/gpt_pydantic_tools-0.2-py3-none-any.whl",
            "yanked": false,
            "yanked_reason": null
        },
        {
            "comment_text": "",
            "digests": {
                "blake2b_256": "1eb1bc022cc16e6f3a3b46ae59d52ff996ea63e1527375755f94641a311633a2",
                "md5": "ae39800896abd8847ad9a1371627214b",
                "sha256": "79a2258190c03c84624b494d21bab5ab66fdac7a86adb9cdbb66de22e4383918"
            },
            "downloads": -1,
            "filename": "gpt_pydantic_tools-0.2.tar.gz",
            "has_sig": false,
            "md5_digest": "ae39800896abd8847ad9a1371627214b",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": null,
            "size": 4433,
            "upload_time": "2024-05-19T17:41:52",
            "upload_time_iso_8601": "2024-05-19T17:41:52.631469Z",
            "url": "https://files.pythonhosted.org/packages/1e/b1/bc022cc16e6f3a3b46ae59d52ff996ea63e1527375755f94641a311633a2/gpt_pydantic_tools-0.2.tar.gz",
            "yanked": false,
            "yanked_reason": null
        }
    ],
    "upload_time": "2024-05-19 17:41:52",
    "github": true,
    "gitlab": false,
    "bitbucket": false,
    "codeberg": false,
    "github_user": "carlosplanchon",
    "github_project": "gpt_pydantic_tools",
    "travis_ci": false,
    "coveralls": false,
    "github_actions": false,
    "requirements": [],
    "lcname": "gpt-pydantic-tools"
}
        
Elapsed time: 0.33202s