# Prompflow-rag
The promptflow-rag package is part of the Promptflow sdk and contains functionality for building indexes locally
## Getting started
### Prerequisites
- Python 3.7 or later is required to use this package.
- You must have an [Azure subscription][azure_subscription].
- An [Azure Machine Learning Workspace][workspace].
- An [Azure AI Studio project][ai_project].
### Install the package
Install the Azure AI generative package for Python with pip:
```
pip install prompflow-rag
```
## Usage
### Create index locally
Users can create an index on their local machine from local source files using the `build_index` method. Given below is a sample.
```python
from promptflow.rag.resources import LocalSource, AzureAISearchConfig, EmbeddingsModelConfig
from promptflow.rag import build_index
# build the index
ai_search_index_path=build_index(
name=index_name, # name of your index
vector_store="azure_ai_search", # the type of vector store - in this case it is Azure AI Search.
embeddings_model_config=EmbeddingsModelConfig(
embeddings_model=f"azure_open_ai://deployment/{embedding_model_deployment}/model/{embedding_model_name}"
)
input_source=LocalSource(input_data="data/product-info/"), # the location of your file/folders
index_config=AzureAISearchConfig(
ai_search_index_name=ai_search_index_name # the name of the index store inside the azure ai search service
)
)
```
The build index will return the path where the index was created.
## Examples
# TODO: add link to sample notebooks
<!-- LINKS -->
[ai_project]: https://aka.ms/azureaistudio
[azure_subscription]: https://azure.microsoft.com/free/
[workspace]: https://docs.microsoft.com/azure/machine-learning/concept-workspace
# promptflow-rag package
Please insert change log into "Next Release" ONLY.
## Next release
## 0.1.0
- ConnectionConfig parameter rename
## 0.0.1
- Introduced package
Raw data
{
"_id": null,
"home_page": "https://github.com/microsoft/promptflow",
"name": "promptflow-rag",
"maintainer": null,
"docs_url": null,
"requires_python": "<4.0,>=3.8",
"maintainer_email": null,
"keywords": null,
"author": "Microsoft Corporation",
"author_email": "aml-pt-eng@microsoft.com",
"download_url": null,
"platform": null,
"description": "# Prompflow-rag\r\n\r\nThe promptflow-rag package is part of the Promptflow sdk and contains functionality for building indexes locally\r\n\r\n## Getting started\r\n\r\n### Prerequisites\r\n\r\n- Python 3.7 or later is required to use this package.\r\n- You must have an [Azure subscription][azure_subscription].\r\n- An [Azure Machine Learning Workspace][workspace].\r\n- An [Azure AI Studio project][ai_project].\r\n\r\n### Install the package\r\n\r\nInstall the Azure AI generative package for Python with pip:\r\n\r\n```\r\npip install prompflow-rag\r\n```\r\n\r\n## Usage\r\n\r\n### Create index locally\r\n\r\nUsers can create an index on their local machine from local source files using the `build_index` method. Given below is a sample.\r\n\r\n```python\r\nfrom promptflow.rag.resources import LocalSource, AzureAISearchConfig, EmbeddingsModelConfig\r\nfrom promptflow.rag import build_index\r\n\r\n# build the index\r\nai_search_index_path=build_index(\r\n name=index_name, # name of your index\r\n vector_store=\"azure_ai_search\", # the type of vector store - in this case it is Azure AI Search.\r\n embeddings_model_config=EmbeddingsModelConfig(\r\n embeddings_model=f\"azure_open_ai://deployment/{embedding_model_deployment}/model/{embedding_model_name}\"\r\n )\r\n input_source=LocalSource(input_data=\"data/product-info/\"), # the location of your file/folders\r\n index_config=AzureAISearchConfig(\r\n ai_search_index_name=ai_search_index_name # the name of the index store inside the azure ai search service\r\n )\r\n)\r\n```\r\n\r\nThe build index will return the path where the index was created.\r\n\r\n## Examples\r\n\r\n# TODO: add link to sample notebooks\r\n\r\n<!-- LINKS -->\r\n\r\n[ai_project]: https://aka.ms/azureaistudio\r\n[azure_subscription]: https://azure.microsoft.com/free/\r\n[workspace]: https://docs.microsoft.com/azure/machine-learning/concept-workspace\r\n\r\n\r\n# promptflow-rag package\r\n\r\nPlease insert change log into \"Next Release\" ONLY.\r\n\r\n## Next release\r\n\r\n## 0.1.0\r\n\r\n- ConnectionConfig parameter rename\r\n\r\n## 0.0.1\r\n\r\n- Introduced package\r\n",
"bugtrack_url": null,
"license": null,
"summary": "Prompt flow index build",
"version": "0.1.0",
"project_urls": {
"Bug Reports": "https://github.com/microsoft/promptflow/issues",
"Homepage": "https://github.com/microsoft/promptflow",
"Source": "https://github.com/microsoft/promptflow"
},
"split_keywords": [],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "d97fdd001f98bb56681824953f70a3e7b2cb8e61b95ddb98d7dde553ff4fd200",
"md5": "db7911ed5f7cece8c5861a1335f68216",
"sha256": "86eb2ed880a1dcc0ad2e3a6dce4583c5ecadc4638f7af22197e225e0cdfb8703"
},
"downloads": -1,
"filename": "promptflow_rag-0.1.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "db7911ed5f7cece8c5861a1335f68216",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "<4.0,>=3.8",
"size": 13379,
"upload_time": "2024-05-10T23:25:35",
"upload_time_iso_8601": "2024-05-10T23:25:35.383716Z",
"url": "https://files.pythonhosted.org/packages/d9/7f/dd001f98bb56681824953f70a3e7b2cb8e61b95ddb98d7dde553ff4fd200/promptflow_rag-0.1.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-05-10 23:25:35",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "microsoft",
"github_project": "promptflow",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "promptflow-rag"
}