gerri 1.0.0

Giant Eagle Retrieval and Response Interface (Gerri) Library

Gerri Library

The Gerri Library (Giant Eagle Retrieval and Response Interface) is a comprehensive Python package designed for managing and orchestrating various functionalities such as database interactions, API management, large language model integration, and more.

Features

• Blob Manager: Manage Azure Blob Storage for file and folder operations.
• API Manager: Handle API-related functionalities (future development).
• Content Filters: Implement and manage content filtering strategies.
• Database Manager: Interact with Cosmos DB and manage database operations.
• LLM Manager: Integrate with OpenAI, Hugging Face, and other LLM APIs.
• Search Manager: Utilize Azure Search for content search capabilities.
• RAG Manager: Manage Retrieval-Augmented Generation workflows.
• Summary Tag Manager: Provide summary tagging functionality.
• Utilities: General utility functions for common tasks.

Installation

To install the Gerri Library, use the following command:

pip install gerri

For development purposes, install the package in editable mode:

pip install -e .

Requirements

Ensure the dependencies listed in gerri/requirements.txt are installed:

pip install -r gerri/requirements.txt

Usage

Here are some examples of how to use the library:

Blob Manager

from gerri.blob_manager import blob_storage_manager

manager = blob_storage_manager.BlobStorageManager(config={
    "connection_string": "your_connection_string",
    "container_name": "your_container_name"
})
manager.upload_file("example.txt", "blob_name")

LLM Manager (OpenAI Example)

from gerri.llm_manager.openai_manager import openai_manager

openai_instance = openai_manager.OpenAIManager(api_key="your_openai_api_key")
response = openai_instance.generate_text(prompt="Hello, world!")
print(response)

Documentation

Full documentation is available in the docs/ directory or online (if hosted on platforms like Read the Docs).

Project Structure

• blob_manager/: Azure Blob Storage management.

• api_manager/: API management utilities (future development).

• content_filters/: Content filtering tools.

• database_manager/: Database management with Cosmos DB.

• llm_manager/: Large language model integrations, including (list):

  • OpenAI
  • Hugging Face
  • Ollama Manager

• rag_manager/: Retrieval-Augmented Generation workflows.

• search_manager/: Azure Search integration.

• summary_tag_manager/: Summary tagging.

• utilities/: General utility functions.

Contributing

Contributions are welcome! Please fork the repository and create a pull request for any enhancements or bug fixes.

License

This project is licensed under the MIT License.

Environment Variables Configuration

This file provides a description of the environment variables required for the Recipe Generator project. These variables are stored in the .env file and are used to configure various services and APIs.

Environment Variables

OpenAI Configuration

• OPENAI_ENDPOINT: The endpoint for the OpenAI service.
• OPENAI_API_KEY: The API key for accessing the OpenAI service.
• OPENAI_DEPLOYMENT_ID: The deployment ID for the OpenAI model.
• OPENAI_API_VERSION: The version of the OpenAI API to use.
• TEMPERATURE: The temperature setting for the OpenAI model.

Image API Configuration

• IMAGE_API_ENDPOINT: The endpoint for the image generation API.
• IMAGE_API_KEY: The API key for accessing the image generation API.
• IMAGE_DEPLOYMENT_ID: The deployment ID for the image generation model.

Content Safety Configuration

• CONTENT_SAFETY_ENDPOINT: The endpoint for the content safety API.
• CONTENT_SAFETY_KEY: The API key for accessing the content safety API.

Cosmos DB Configuration

• COSMOS_DB_ENDPOINT: The endpoint for the Cosmos DB service.
• COSMOS_DB_KEY: The key for accessing the Cosmos DB service.
• COSMOS_DB_DATABASE_NAME: The name of the Cosmos DB database.
• COSMOS_DB_CONTAINER_NAME: The name of the Cosmos DB container.
• COSMOS_DB_PARTITION_KEY: The partition key for the Cosmos DB container.

Recipe Generation Prompt

• RECIPE_GENERATION_PROMPT: The prompt used for generating recipes. This should be a JSON string with specific guidelines for the recipe generation.

Azure Blob Storage Configuration

• BLOB_CONNECTION_STRING: The connection string for accessing Azure Blob Storage.
• BLOB_CONTAINER_NAME: The name of the blob container where recipe files are stored.
• BLOB_RECIPE_FOLDER: The folder within the blob container where recipe files are stored.

Azure Search Configuration

• AZURE_SEARCH_ENDPOINT: The endpoint for the Azure Search service.
• AZURE_SEARCH_KEY: The API key for accessing the Azure Search service.
• AZURE_SEARCH_INDEX_NAME: The name of the Azure Search index.
• AZURE_SEARCH_INDEXER_NAME: The name of the Azure Search indexer.
• AZURE_SEARCH_INDEXER_INTERVAL_IN_SECONDS: The interval in seconds for the Azure Search indexer to run.

Form Recognizer Configuration

• FORM_RECOGNIZER_ENDPOINT: The endpoint for the Azure Form Recognizer service.
• FORM_RECOGNIZER_KEY: The API key for accessing the Azure Form Recognizer service.

System Prompt

• SYSTEM_PROMPT: The system prompt used for generating recipes. This should be a JSON string with specific guidelines for the recipe generation.

LLM Type

• LLM_TYPE: The type of language model to use (e.g., openai or ollama). ollama will use local LLM and will require OLLAMA to be installed on the machine running the code.
• OLLAMA_MODEL: specify the Ollama model to be used like LLAMA3.1 or LLAMA3.2