| Name | llm-loop JSON |
| Version |
0.2
JSON |
| download |
| home_page | https://github.com/chigwell/llm-loop |
| Summary | A utility package for querying language models with pattern matching and retry logic |
| upload_time | 2023-12-10 17:14:37 |
| maintainer | |
| docs_url | None |
| author | Evgenii Evstafev |
| requires_python | |
| license | |
| keywords |
|
| VCS |
 |
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
# llm-loop
## Overview
`llm-loop` is a Python package designed to simplify the process of querying language models (like GPT or similar models) until a response matching a specified pattern is obtained or a maximum number of attempts is reached. This is particularly useful when working with AI models in scenarios where a specific format of response is required.
## Installation
```bash
pip install llm-loop
```
This will install the necessary Python packages, including `ctransformers` and any other dependencies.
## Usage
Here's a basic example of how to use `llm-loop`:
1. **Import the necessary modules:**
```python
import os
from ctransformers import AutoModelForCausalLM, AutoTokenizer
from llm_loop.main import LLMLoop
```
2. **Initialize the model with custom parameters:**
```python
model_name = "YourModelName"
model_file = "YourModelFileName"
start_dir = '/path/to/your/model'
model_path = f"{start_dir}/{model_file}"
llm = AutoModelForCausalLM.from_pretrained(model_name, model_file=model_path, model_type='YourModelType', gpu_layers=YourGPULayers)
```
3. **Create an instance of LLMLoop and query the model:**
```python
loop = LLMLoop(llm, 10) # 10 is the maximum number of attempts
prompt = "Your prompt here"
pattern = r'Your regex pattern here'
response = loop.query_llm(prompt=prompt, pattern=pattern)
print("Response:", response)
```
## Contributing
Contributions to `llm-loop` are welcome! Please feel free to submit pull requests or open issues to suggest improvements or add new features.
## License
MIT.
Raw data
{
"_id": null,
"home_page": "https://github.com/chigwell/llm-loop",
"name": "llm-loop",
"maintainer": "",
"docs_url": null,
"requires_python": "",
"maintainer_email": "",
"keywords": "",
"author": "Evgenii Evstafev",
"author_email": "chigwel@gmail.com",
"download_url": "https://files.pythonhosted.org/packages/47/92/147b3760a05e28f00d715425f5dcbff0b815212507746ff324078ad3ae6b/llm-loop-0.2.tar.gz",
"platform": null,
"description": "# llm-loop\n\n## Overview\n\n`llm-loop` is a Python package designed to simplify the process of querying language models (like GPT or similar models) until a response matching a specified pattern is obtained or a maximum number of attempts is reached. This is particularly useful when working with AI models in scenarios where a specific format of response is required.\n\n## Installation\n\n```bash\npip install llm-loop\n```\n\n This will install the necessary Python packages, including `ctransformers` and any other dependencies.\n\n## Usage\n\nHere's a basic example of how to use `llm-loop`:\n\n1. **Import the necessary modules:**\n ```python\n import os\n from ctransformers import AutoModelForCausalLM, AutoTokenizer\n from llm_loop.main import LLMLoop\n ```\n\n2. **Initialize the model with custom parameters:**\n ```python\n model_name = \"YourModelName\"\n model_file = \"YourModelFileName\"\n start_dir = '/path/to/your/model'\n model_path = f\"{start_dir}/{model_file}\"\n\n llm = AutoModelForCausalLM.from_pretrained(model_name, model_file=model_path, model_type='YourModelType', gpu_layers=YourGPULayers)\n ```\n\n3. **Create an instance of LLMLoop and query the model:**\n ```python\n loop = LLMLoop(llm, 10) # 10 is the maximum number of attempts\n\n prompt = \"Your prompt here\"\n pattern = r'Your regex pattern here'\n\n response = loop.query_llm(prompt=prompt, pattern=pattern)\n\n print(\"Response:\", response)\n ```\n\n## Contributing\n\nContributions to `llm-loop` are welcome! Please feel free to submit pull requests or open issues to suggest improvements or add new features.\n\n## License\n\nMIT.\n",
"bugtrack_url": null,
"license": "",
"summary": "A utility package for querying language models with pattern matching and retry logic",
"version": "0.2",
"project_urls": {
"Homepage": "https://github.com/chigwell/llm-loop"
},
"split_keywords": [],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "d8fa75c25eaa96461f9483f884fcd4e38c652a5d6de452aa35ca523664d2e317",
"md5": "900f49d6462e130f59795f0e6896c288",
"sha256": "0f33cf89d7d08f0cbf13c4b020126ed1ac50f32c408fa78d3fc3be411045fe3e"
},
"downloads": -1,
"filename": "llm_loop-0.2-py3-none-any.whl",
"has_sig": false,
"md5_digest": "900f49d6462e130f59795f0e6896c288",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": null,
"size": 1850,
"upload_time": "2023-12-10T17:14:35",
"upload_time_iso_8601": "2023-12-10T17:14:35.187093Z",
"url": "https://files.pythonhosted.org/packages/d8/fa/75c25eaa96461f9483f884fcd4e38c652a5d6de452aa35ca523664d2e317/llm_loop-0.2-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "4792147b3760a05e28f00d715425f5dcbff0b815212507746ff324078ad3ae6b",
"md5": "9f702aa72e93627a3f21e1149138d034",
"sha256": "02eadb0b7ed176210ad6cccf3b32864a0b65eef4e4a541386c10c9076d1bd9fd"
},
"downloads": -1,
"filename": "llm-loop-0.2.tar.gz",
"has_sig": false,
"md5_digest": "9f702aa72e93627a3f21e1149138d034",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 1882,
"upload_time": "2023-12-10T17:14:37",
"upload_time_iso_8601": "2023-12-10T17:14:37.395034Z",
"url": "https://files.pythonhosted.org/packages/47/92/147b3760a05e28f00d715425f5dcbff0b815212507746ff324078ad3ae6b/llm-loop-0.2.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-12-10 17:14:37",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "chigwell",
"github_project": "llm-loop",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"lcname": "llm-loop"
}
