| Name | neuraltrust JSON |
| Version |
1.2.0
JSON |
| download |
| home_page | None |
| Summary | A Python library for NeuralTrust API |
| upload_time | 2024-10-17 13:11:33 |
| maintainer | None |
| docs_url | None |
| author | NeuralTrust |
| requires_python | <4.0,>=3.10 |
| license | Commercial |
| keywords |
|
| VCS |
|
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
# NeuralTrust Python SDK
The NeuralTrust Python SDK provides a convenient way to interact with the NeuralTrust API for tracing, evaluation sets, knowledge bases, and testsets.
## Installation
You can install the NeuralTrust Python SDK using pip:
```bash
pip install neuraltrust
```
## Usage
To use the NeuralTrust Python SDK, you need to initialize the client with your API key:
```python
from neuraltrust import NeuralTrust
# Initialize the client with your API key
client = NeuralTrust(api_key="your_api_key_here")
# Optionally, you can specify a custom base URL and SDK version
client = NeuralTrust(api_key="your_api_key_here", base_url="https://custom.api.url", sdk_version="v2")
```
### Tracing
```python
# Create a trace
trace = client.trace(trace_id="trace_1234", conversation_id="conversation_1234", session_id="session_1234", channel_id="channel_1234")
# Add events to the trace
trace.retrieval("What's the weather like today?")
# Rest of your code and the end you call the end method
trace.end([{"chunk": "The weather today is sunny with a high of 75°F.", "score": 0.95}])
# There is another method to send the trace in an atomic way
trace.send(
event_type=EventType.RETRIEVAL,
input="What's the weather like today?",
output=[{"chunk": "The weather today is sunny with a high of 75°F.", "score": 0.95}],
latency=100
)
```
### Evaluation Sets
```python
# Run evaluation set
eval_set = client.run_evaluation_set(id="eval_set_id")
# Create an evaluation set
eval_set = client.create_evaluation_set(name="My Eval Set", description="A test evaluation set")
# Get an evaluation set
eval_set = client.get_evaluation_set(id="eval_set_id")
# Delete an evaluation set
client.delete_evaluation_set(id="eval_set_id")
```
### Knowledge Bases
```python
# Create a knowledge base
kb = client.create_knowledge_base(type="upstash", credentials={"api_key": "your_doc_api_key"})
# Get a knowledge base
kb = client.get_knowledge_base(id="kb_id")
# Delete a knowledge base
client.delete_knowledge_base(id="kb_id")
```
### Testsets
```python
# Create a testset
testset = client.create_testset(name="My Testset", type="adversarial", evaluation_set_id="eval_set_id", knowledge_base_id="kb_id", num_questions=10)
# Get a testset
testset = client.get_testset(id="testset_id")
# Delete a testset
client.delete_testset(id="testset_id")
```
## Configuration
You can configure the SDK using environment variables:
- `NEURALTRUST_API_KEY`: Your NeuralTrust API key
- `NEURALTRUST_BASE_URL`: Custom base URL for the API (optional)
## Advanced Usage
### Custom Metadata and User Information
```python
from neuraltrust import User, Metadata
user = User(id="user123", name="John Doe")
metadata = Metadata(app_version="1.0.0", platform="web")
trace = client.trace(user=user, metadata=metadata)
```
### Asynchronous Tracing
The SDK uses a `ThreadPoolExecutor` for asynchronous tracing. You can adjust the number of workers:
```python
client = NeuralTrust(api_key="your_api_key_here", max_workers=10)
```
## Error Handling
The SDK will raise exceptions for API errors. Make sure to handle these appropriately in your application.
## Contributing
Contributions to the NeuralTrust Python SDK are welcome! Please refer to the contribution guidelines for more information.
## License
This SDK is distributed under the [MIT License](LICENSE).
Raw data
{
"_id": null,
"home_page": null,
"name": "neuraltrust",
"maintainer": null,
"docs_url": null,
"requires_python": "<4.0,>=3.10",
"maintainer_email": null,
"keywords": null,
"author": "NeuralTrust",
"author_email": "support@neuraltrust.ai",
"download_url": "https://files.pythonhosted.org/packages/fd/c3/ce0ee4b787b1a201be2b78329590baebe95fc4a87917eeaab4c50083f1f1/neuraltrust-1.2.0.tar.gz",
"platform": null,
"description": "# NeuralTrust Python SDK\n\nThe NeuralTrust Python SDK provides a convenient way to interact with the NeuralTrust API for tracing, evaluation sets, knowledge bases, and testsets.\n\n## Installation\n\nYou can install the NeuralTrust Python SDK using pip:\n\n```bash\npip install neuraltrust\n```\n\n## Usage\n\nTo use the NeuralTrust Python SDK, you need to initialize the client with your API key:\n\n```python\nfrom neuraltrust import NeuralTrust\n\n# Initialize the client with your API key\nclient = NeuralTrust(api_key=\"your_api_key_here\")\n\n# Optionally, you can specify a custom base URL and SDK version\nclient = NeuralTrust(api_key=\"your_api_key_here\", base_url=\"https://custom.api.url\", sdk_version=\"v2\")\n```\n\n### Tracing\n\n```python\n# Create a trace\ntrace = client.trace(trace_id=\"trace_1234\", conversation_id=\"conversation_1234\", session_id=\"session_1234\", channel_id=\"channel_1234\")\n\n# Add events to the trace\ntrace.retrieval(\"What's the weather like today?\")\n# Rest of your code and the end you call the end method\ntrace.end([{\"chunk\": \"The weather today is sunny with a high of 75\u00b0F.\", \"score\": 0.95}])\n\n# There is another method to send the trace in an atomic way\ntrace.send(\n event_type=EventType.RETRIEVAL, \n input=\"What's the weather like today?\", \n output=[{\"chunk\": \"The weather today is sunny with a high of 75\u00b0F.\", \"score\": 0.95}], \n latency=100\n)\n```\n\n### Evaluation Sets\n\n```python\n# Run evaluation set\neval_set = client.run_evaluation_set(id=\"eval_set_id\")\n\n# Create an evaluation set\neval_set = client.create_evaluation_set(name=\"My Eval Set\", description=\"A test evaluation set\")\n\n# Get an evaluation set\neval_set = client.get_evaluation_set(id=\"eval_set_id\")\n\n# Delete an evaluation set\nclient.delete_evaluation_set(id=\"eval_set_id\")\n```\n\n### Knowledge Bases\n\n```python\n# Create a knowledge base\nkb = client.create_knowledge_base(type=\"upstash\", credentials={\"api_key\": \"your_doc_api_key\"})\n\n# Get a knowledge base\nkb = client.get_knowledge_base(id=\"kb_id\")\n\n# Delete a knowledge base\nclient.delete_knowledge_base(id=\"kb_id\")\n```\n\n### Testsets\n\n```python\n# Create a testset\ntestset = client.create_testset(name=\"My Testset\", type=\"adversarial\", evaluation_set_id=\"eval_set_id\", knowledge_base_id=\"kb_id\", num_questions=10)\n\n# Get a testset\ntestset = client.get_testset(id=\"testset_id\")\n\n# Delete a testset\nclient.delete_testset(id=\"testset_id\")\n```\n\n## Configuration\n\nYou can configure the SDK using environment variables:\n\n- `NEURALTRUST_API_KEY`: Your NeuralTrust API key\n- `NEURALTRUST_BASE_URL`: Custom base URL for the API (optional)\n\n## Advanced Usage\n\n### Custom Metadata and User Information\n\n```python\nfrom neuraltrust import User, Metadata\n\nuser = User(id=\"user123\", name=\"John Doe\")\nmetadata = Metadata(app_version=\"1.0.0\", platform=\"web\")\n\ntrace = client.trace(user=user, metadata=metadata)\n```\n\n### Asynchronous Tracing\n\nThe SDK uses a `ThreadPoolExecutor` for asynchronous tracing. You can adjust the number of workers:\n\n```python\nclient = NeuralTrust(api_key=\"your_api_key_here\", max_workers=10)\n```\n\n## Error Handling\n\nThe SDK will raise exceptions for API errors. Make sure to handle these appropriately in your application.\n\n## Contributing\n\nContributions to the NeuralTrust Python SDK are welcome! Please refer to the contribution guidelines for more information.\n\n## License\n\nThis SDK is distributed under the [MIT License](LICENSE).\n\n\n",
"bugtrack_url": null,
"license": "Commercial",
"summary": "A Python library for NeuralTrust API",
"version": "1.2.0",
"project_urls": null,
"split_keywords": [],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "111b551917b24e70066a1ea439e7f93e6385d85ec7f78b30292d878c350ba2f2",
"md5": "0f58b4f83da0847cf6d465ae4fdbda3a",
"sha256": "9f51c7c4262e64241122580a9e574d7a2891f95406bc98743da2776e0c226c7d"
},
"downloads": -1,
"filename": "neuraltrust-1.2.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "0f58b4f83da0847cf6d465ae4fdbda3a",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "<4.0,>=3.10",
"size": 40199,
"upload_time": "2024-10-17T13:11:32",
"upload_time_iso_8601": "2024-10-17T13:11:32.302621Z",
"url": "https://files.pythonhosted.org/packages/11/1b/551917b24e70066a1ea439e7f93e6385d85ec7f78b30292d878c350ba2f2/neuraltrust-1.2.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "fdc3ce0ee4b787b1a201be2b78329590baebe95fc4a87917eeaab4c50083f1f1",
"md5": "365e45d26c36597b5c42b2d77b11c01d",
"sha256": "6a83c1dbe5dc433a9e9a87ad25919ff570d4b899961a06791bce952f6bdf23e2"
},
"downloads": -1,
"filename": "neuraltrust-1.2.0.tar.gz",
"has_sig": false,
"md5_digest": "365e45d26c36597b5c42b2d77b11c01d",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "<4.0,>=3.10",
"size": 24495,
"upload_time": "2024-10-17T13:11:33",
"upload_time_iso_8601": "2024-10-17T13:11:33.759229Z",
"url": "https://files.pythonhosted.org/packages/fd/c3/ce0ee4b787b1a201be2b78329590baebe95fc4a87917eeaab4c50083f1f1/neuraltrust-1.2.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-10-17 13:11:33",
"github": false,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"lcname": "neuraltrust"
}
