jaas-ai


Namejaas-ai JSON
Version 0.1.9 PyPI version JSON
download
home_pageNone
SummaryPython SDK for JaaS AI evaluation API
upload_time2025-09-08 18:05:34
maintainerNone
docs_urlNone
authorNone
requires_python>=3.7
licenseNone
keywords ai evaluation api sdk machine learning llm
VCS
bugtrack_url
requirements No requirements were recorded.
Travis-CI No Travis.
coveralls test coverage No coveralls.
            # JaaS API Python SDK

[![PyPI version](https://badge.fury.io/py/jaas_ai.svg)](https://badge.fury.io/py/jaas_ai)
[![Python Support](https://img.shields.io/pypi/pyversions/jaas_ai.svg)](https://pypi.org/project/jaas_ai/)
[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)

A Python SDK for interacting with the JaaS AI evaluation API. This library provides a simple and intuitive interface for evaluating AI-generated responses against various criteria.

Visit jaas-ai.net for more information.
Note: a JaaS AI API key is required. 

## Features

- 🚀 Simple and intuitive API
- 🔧 Support for multiple evaluation types (S, C, D, V), with multiple citeria
- 📊 Comprehensive evaluation criteria support including Hallucination Detection
- 🛡️ Robust error handling
- 📝 Type hints for better IDE support
- 🐍 Python 3.7+ support

## Installation

### From PyPI (recommended)

```bash
pip install jaas_ai
```


## Quick Start

```python
from jaas_ai import jaas_client

# Initialize the client with your API key
client = jaas_client(api_key="your-api-key-here")

# Evaluate an AI response
result = client.evaluate(
    question="What is the capital of France?",
    answer="The capital of France is Paris.",
    evaluation_criteria=["accuracy", "completeness", "clarity"],
    eval_type="S"
)

print(result)
```

## API Reference

### jaas_client

The main client class for interacting with the JaaS API.

#### Constructor

```python
jaas_client(api_key: str)
```

- `api_key` (str): Your JaaS API key

#### Methods

##### evaluate()

Submit an evaluation request to the JaaS API.

```python
def evaluate(
    self,
    question: str,
    answer: str,
    evaluation_criteria: List[str],
    eval_type: str = "S",
    ground_truth_answer: Optional[str] = None,
    context: Optional[str] = None,
    cohort: Optional[str] = None
) -> Dict
```

**Parameters:**

- `question` (str): The input question
- `answer` (str): The AI-generated answer to evaluate
- `evaluation_criteria` (List[str]): List of criteria to evaluate against
- `eval_type` (str, optional): Evaluation type. Options:
- `ground_truth_answer` (str, optional): Reference answer for comparison
- `context` (str, optional): Additional context for evaluation
- `cohort` (str, optional): Cohort name for grouping evaluations

**Returns:**

- `Dict`: Evaluation results from the API

**Raises:**

- `requests.exceptions.ConnectionError`: If unable to connect to the API
- `requests.exceptions.Timeout`: If the request times out
- `requests.exceptions.RequestException`: For other request-related errors

## License

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

## Support

- 📧 Email: info@jaas-ai.com
- 📖 Documentation: [https://jaas-ai.net ]

## Changelog

### v0.1.5 (2025-09-08)
- Improved error handling with cleaner error messages
- Better HTTP error reporting from server responses

### v0.1.4 (2025-09-08)
- Simplified use and API call
- Improved evaluation functionality
- Support for multiple evaluation types
- Comprehensive error handling


            

Raw data

            {
    "_id": null,
    "home_page": null,
    "name": "jaas-ai",
    "maintainer": null,
    "docs_url": null,
    "requires_python": ">=3.7",
    "maintainer_email": "JaaS <info@jaas-ai.com>",
    "keywords": "ai, evaluation, api, sdk, machine learning, llm",
    "author": null,
    "author_email": "JaaS <info@jaas-ai.com>",
    "download_url": "https://files.pythonhosted.org/packages/d6/42/d03c6a18553d40b906d6f265bf90d096e38a58063f8e3afb700b1082c68e/jaas_ai-0.1.9.tar.gz",
    "platform": null,
    "description": "# JaaS API Python SDK\r\n\r\n[![PyPI version](https://badge.fury.io/py/jaas_ai.svg)](https://badge.fury.io/py/jaas_ai)\r\n[![Python Support](https://img.shields.io/pypi/pyversions/jaas_ai.svg)](https://pypi.org/project/jaas_ai/)\r\n[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)\r\n\r\nA Python SDK for interacting with the JaaS AI evaluation API. This library provides a simple and intuitive interface for evaluating AI-generated responses against various criteria.\r\n\r\nVisit jaas-ai.net for more information.\r\nNote: a JaaS AI API key is required. \r\n\r\n## Features\r\n\r\n- \ud83d\ude80 Simple and intuitive API\r\n- \ud83d\udd27 Support for multiple evaluation types (S, C, D, V), with multiple citeria\r\n- \ud83d\udcca Comprehensive evaluation criteria support including Hallucination Detection\r\n- \ud83d\udee1\ufe0f Robust error handling\r\n- \ud83d\udcdd Type hints for better IDE support\r\n- \ud83d\udc0d Python 3.7+ support\r\n\r\n## Installation\r\n\r\n### From PyPI (recommended)\r\n\r\n```bash\r\npip install jaas_ai\r\n```\r\n\r\n\r\n## Quick Start\r\n\r\n```python\r\nfrom jaas_ai import jaas_client\r\n\r\n# Initialize the client with your API key\r\nclient = jaas_client(api_key=\"your-api-key-here\")\r\n\r\n# Evaluate an AI response\r\nresult = client.evaluate(\r\n    question=\"What is the capital of France?\",\r\n    answer=\"The capital of France is Paris.\",\r\n    evaluation_criteria=[\"accuracy\", \"completeness\", \"clarity\"],\r\n    eval_type=\"S\"\r\n)\r\n\r\nprint(result)\r\n```\r\n\r\n## API Reference\r\n\r\n### jaas_client\r\n\r\nThe main client class for interacting with the JaaS API.\r\n\r\n#### Constructor\r\n\r\n```python\r\njaas_client(api_key: str)\r\n```\r\n\r\n- `api_key` (str): Your JaaS API key\r\n\r\n#### Methods\r\n\r\n##### evaluate()\r\n\r\nSubmit an evaluation request to the JaaS API.\r\n\r\n```python\r\ndef evaluate(\r\n    self,\r\n    question: str,\r\n    answer: str,\r\n    evaluation_criteria: List[str],\r\n    eval_type: str = \"S\",\r\n    ground_truth_answer: Optional[str] = None,\r\n    context: Optional[str] = None,\r\n    cohort: Optional[str] = None\r\n) -> Dict\r\n```\r\n\r\n**Parameters:**\r\n\r\n- `question` (str): The input question\r\n- `answer` (str): The AI-generated answer to evaluate\r\n- `evaluation_criteria` (List[str]): List of criteria to evaluate against\r\n- `eval_type` (str, optional): Evaluation type. Options:\r\n- `ground_truth_answer` (str, optional): Reference answer for comparison\r\n- `context` (str, optional): Additional context for evaluation\r\n- `cohort` (str, optional): Cohort name for grouping evaluations\r\n\r\n**Returns:**\r\n\r\n- `Dict`: Evaluation results from the API\r\n\r\n**Raises:**\r\n\r\n- `requests.exceptions.ConnectionError`: If unable to connect to the API\r\n- `requests.exceptions.Timeout`: If the request times out\r\n- `requests.exceptions.RequestException`: For other request-related errors\r\n\r\n## License\r\n\r\nThis project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.\r\n\r\n## Support\r\n\r\n- \ud83d\udce7 Email: info@jaas-ai.com\r\n- \ud83d\udcd6 Documentation: [https://jaas-ai.net ]\r\n\r\n## Changelog\r\n\r\n### v0.1.5 (2025-09-08)\r\n- Improved error handling with cleaner error messages\r\n- Better HTTP error reporting from server responses\r\n\r\n### v0.1.4 (2025-09-08)\r\n- Simplified use and API call\r\n- Improved evaluation functionality\r\n- Support for multiple evaluation types\r\n- Comprehensive error handling\r\n\r\n",
    "bugtrack_url": null,
    "license": null,
    "summary": "Python SDK for JaaS AI evaluation API",
    "version": "0.1.9",
    "project_urls": null,
    "split_keywords": [
        "ai",
        " evaluation",
        " api",
        " sdk",
        " machine learning",
        " llm"
    ],
    "urls": [
        {
            "comment_text": null,
            "digests": {
                "blake2b_256": "42dbbb660489fa634793afeca74de0dac91957075fa04e5a1728187e93ebd7b3",
                "md5": "7bf9692e9e04f2827f1241e351c96851",
                "sha256": "ccc705012145672f314c25b007fef661655e6ef544a47bb66995bc6c124305de"
            },
            "downloads": -1,
            "filename": "jaas_ai-0.1.9-py3-none-any.whl",
            "has_sig": false,
            "md5_digest": "7bf9692e9e04f2827f1241e351c96851",
            "packagetype": "bdist_wheel",
            "python_version": "py3",
            "requires_python": ">=3.7",
            "size": 5519,
            "upload_time": "2025-09-08T18:05:33",
            "upload_time_iso_8601": "2025-09-08T18:05:33.778146Z",
            "url": "https://files.pythonhosted.org/packages/42/db/bb660489fa634793afeca74de0dac91957075fa04e5a1728187e93ebd7b3/jaas_ai-0.1.9-py3-none-any.whl",
            "yanked": false,
            "yanked_reason": null
        },
        {
            "comment_text": null,
            "digests": {
                "blake2b_256": "d642d03c6a18553d40b906d6f265bf90d096e38a58063f8e3afb700b1082c68e",
                "md5": "6d992755dcec0fb0583207812ad6f7af",
                "sha256": "a014bdc31f4daf5733c87719f3ee4e41abc0ddd5b38c65a9b1dff15617e671cb"
            },
            "downloads": -1,
            "filename": "jaas_ai-0.1.9.tar.gz",
            "has_sig": false,
            "md5_digest": "6d992755dcec0fb0583207812ad6f7af",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": ">=3.7",
            "size": 6450,
            "upload_time": "2025-09-08T18:05:34",
            "upload_time_iso_8601": "2025-09-08T18:05:34.632233Z",
            "url": "https://files.pythonhosted.org/packages/d6/42/d03c6a18553d40b906d6f265bf90d096e38a58063f8e3afb700b1082c68e/jaas_ai-0.1.9.tar.gz",
            "yanked": false,
            "yanked_reason": null
        }
    ],
    "upload_time": "2025-09-08 18:05:34",
    "github": false,
    "gitlab": false,
    "bitbucket": false,
    "codeberg": false,
    "lcname": "jaas-ai"
}
        
Elapsed time: 0.44029s