| Name | usellm JSON |
| Version |
0.0.7
JSON |
| download |
| home_page | https://github.com/usellm/usellm-py |
| Summary | Use Large Language Models in Python App |
| upload_time | 2023-06-15 13:44:23 |
| maintainer | |
| docs_url | None |
| author | Siddhant |
| requires_python | |
| license | |
| keywords |
|
| VCS |
 |
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
# useLLM - Use Large Language Models in Python App
The `usellm` Python library enables interaction with a chat-based Large Language Model (LLM) service. This interaction can be used to perform various language-related tasks, like generating chat conversation using the OpenAI API. It's designed as a Python port of the [`usellm`](https://usellm.org) JavaScript library.
## Installation
The library can be installed with `pip`:
```
pip install usellm
```
## Example Usage
Here is a basic usage example:
```python
from usellm import Message, Options, UseLLM
# Initialize the service
service = UseLLM(service_url="https://usellm.org/api/llm")
# Prepare the conversation
messages = [
Message(role="system", content="You are a helpful assistant."),
Message(role="user", content="What can you do for me?"),
]
options = Options(messages=messages)
# Interact with the service
response = service.chat(options)
# Print the assistant's response
print(response.content)
```
The above code will generate a response using the OpenAI ChatGPT API. The service URL "https://usellm.org/api/llm" should be used only for testing.
## Classes and Methods
### 1. `UseLLM` class
The `UseLLM` class provides the interface for interacting with the LLM service.
Methods:
- `__init__(self, service_url: str)`: Initializes a new instance of the `UseLLM` class.
- `chat(self, options: Options) -> Message`: Interacts with the LLM using the provided `Options`, and returns a `Message` instance that represents the LLM's response.
### 2. `Options` class
The `Options` class represents a set of configuration options for a chat interaction with the LLM.
- `messages`: A list of `Message` instances representing the conversation up to the current point.
- `stream`: A boolean indicating if the interaction is a streaming interaction. Note: streaming is currently not supported.
- `template`: A string representing a message template to guide the conversation.
- `inputs`: A dictionary of additional inputs for the conversation.
Methods:
- `__init__(self, messages: Optional[List[Message]] = [], stream: Optional[bool] = None, template: Optional[str] = None, inputs: Optional[dict] = None)`: Initializes a new instance of the `Options` class.
### 3. `Message` class
The `Message` class represents a message in a conversation. It consists of two main attributes:
- `role`: The role of the message sender. Common values could be `system`, `user`, `assistant`.
- `content`: The content of the message.
Methods:
- `__init__(self, role: str, content: str)`: Initializes a new instance of the `Message` class.
- `__repr__(self) -> str`: Returns a string representation of the `Message` instance.
- `__str__(self) -> str`: Returns a string representation of the `Message` instance.
- `to_dict(self) -> dict`: Returns a dictionary representation of the `Message` instance.
- `to_json(self) -> str`: Returns a JSON string representation of the `Message` instance.
## Exceptions
The library raises an `Exception` in the following situations:
- If the `stream` option is set to `True`, because streaming is not currently supported.
- If the HTTP response status code from the LLM service is not 200.
- If the HTTP response from the LLM service contains an "error" field.
- If the HTTP response from the LLM service does not contain a "choices" field.
Please create an issue to report bugs or suggest improvements. Learn more about the original JavaScript library here: https://usellm.org
Raw data
{
"_id": null,
"home_page": "https://github.com/usellm/usellm-py",
"name": "usellm",
"maintainer": "",
"docs_url": null,
"requires_python": "",
"maintainer_email": "",
"keywords": "",
"author": "Siddhant",
"author_email": "siddhant@jovian.com",
"download_url": "https://files.pythonhosted.org/packages/fd/6b/c214151216d75206cf89c42464e249bcf38a86c04289a9ac2c36dc07c47d/usellm-0.0.7.tar.gz",
"platform": null,
"description": "# useLLM - Use Large Language Models in Python App\n\nThe `usellm` Python library enables interaction with a chat-based Large Language Model (LLM) service. This interaction can be used to perform various language-related tasks, like generating chat conversation using the OpenAI API. It's designed as a Python port of the [`usellm`](https://usellm.org) JavaScript library.\n\n## Installation\n\nThe library can be installed with `pip`:\n\n```\npip install usellm\n```\n\n## Example Usage\n\nHere is a basic usage example:\n\n```python\nfrom usellm import Message, Options, UseLLM\n\n# Initialize the service\nservice = UseLLM(service_url=\"https://usellm.org/api/llm\")\n\n# Prepare the conversation\nmessages = [\n Message(role=\"system\", content=\"You are a helpful assistant.\"),\n Message(role=\"user\", content=\"What can you do for me?\"),\n]\noptions = Options(messages=messages)\n\n# Interact with the service\nresponse = service.chat(options)\n\n# Print the assistant's response\nprint(response.content)\n```\n\nThe above code will generate a response using the OpenAI ChatGPT API. The service URL \"https://usellm.org/api/llm\" should be used only for testing.\n\n## Classes and Methods\n\n### 1. `UseLLM` class\n\nThe `UseLLM` class provides the interface for interacting with the LLM service.\n\nMethods:\n- `__init__(self, service_url: str)`: Initializes a new instance of the `UseLLM` class.\n- `chat(self, options: Options) -> Message`: Interacts with the LLM using the provided `Options`, and returns a `Message` instance that represents the LLM's response.\n\n### 2. `Options` class\n\nThe `Options` class represents a set of configuration options for a chat interaction with the LLM.\n\n- `messages`: A list of `Message` instances representing the conversation up to the current point.\n- `stream`: A boolean indicating if the interaction is a streaming interaction. Note: streaming is currently not supported.\n- `template`: A string representing a message template to guide the conversation.\n- `inputs`: A dictionary of additional inputs for the conversation.\n\nMethods:\n- `__init__(self, messages: Optional[List[Message]] = [], stream: Optional[bool] = None, template: Optional[str] = None, inputs: Optional[dict] = None)`: Initializes a new instance of the `Options` class.\n\n### 3. `Message` class\n\nThe `Message` class represents a message in a conversation. It consists of two main attributes:\n\n- `role`: The role of the message sender. Common values could be `system`, `user`, `assistant`.\n- `content`: The content of the message.\n\nMethods:\n- `__init__(self, role: str, content: str)`: Initializes a new instance of the `Message` class.\n- `__repr__(self) -> str`: Returns a string representation of the `Message` instance.\n- `__str__(self) -> str`: Returns a string representation of the `Message` instance.\n- `to_dict(self) -> dict`: Returns a dictionary representation of the `Message` instance.\n- `to_json(self) -> str`: Returns a JSON string representation of the `Message` instance.\n\n\n## Exceptions\n\nThe library raises an `Exception` in the following situations:\n\n- If the `stream` option is set to `True`, because streaming is not currently supported.\n- If the HTTP response status code from the LLM service is not 200.\n- If the HTTP response from the LLM service contains an \"error\" field.\n- If the HTTP response from the LLM service does not contain a \"choices\" field.\n\n\nPlease create an issue to report bugs or suggest improvements. Learn more about the original JavaScript library here: https://usellm.org\n",
"bugtrack_url": null,
"license": "",
"summary": "Use Large Language Models in Python App",
"version": "0.0.7",
"project_urls": {
"Homepage": "https://github.com/usellm/usellm-py"
},
"split_keywords": [],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "5f2d3ff77e73796c80ff199ec37bf4acdb1b9334f6364d9b4597c15abf3c08c8",
"md5": "58dced56dfc53ed959e56d4096b341d2",
"sha256": "6c82c15228bace2f6baacc547aecb74a8b4f60ba18d5975cc667126e7511f9b8"
},
"downloads": -1,
"filename": "usellm-0.0.7-py3-none-any.whl",
"has_sig": false,
"md5_digest": "58dced56dfc53ed959e56d4096b341d2",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": null,
"size": 5457,
"upload_time": "2023-06-15T13:44:20",
"upload_time_iso_8601": "2023-06-15T13:44:20.938478Z",
"url": "https://files.pythonhosted.org/packages/5f/2d/3ff77e73796c80ff199ec37bf4acdb1b9334f6364d9b4597c15abf3c08c8/usellm-0.0.7-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "fd6bc214151216d75206cf89c42464e249bcf38a86c04289a9ac2c36dc07c47d",
"md5": "84eede00904ec4e2a1dbc998b22c2d0e",
"sha256": "008e2f02ae03aa37e149e3d3ab92ec8fd72a335a6aedb7c92d678fc9a7a16cf4"
},
"downloads": -1,
"filename": "usellm-0.0.7.tar.gz",
"has_sig": false,
"md5_digest": "84eede00904ec4e2a1dbc998b22c2d0e",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 5140,
"upload_time": "2023-06-15T13:44:23",
"upload_time_iso_8601": "2023-06-15T13:44:23.159889Z",
"url": "https://files.pythonhosted.org/packages/fd/6b/c214151216d75206cf89c42464e249bcf38a86c04289a9ac2c36dc07c47d/usellm-0.0.7.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-06-15 13:44:23",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "usellm",
"github_project": "usellm-py",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"lcname": "usellm"
}
