Azure OpenAI Embeddings QnA

Simple web application for creating Azure OpenAI embeddings vectors, retrieve the most relevant document, and apply QnA on it.

Application Settings

App Setting	Value	Note
engines	davinci-002,text-curie-001,text-babbage-001,text-ada-001	Engines deployed in your Azure OpenAI Resource
embeddings_engines	text-search-davinci-doc-001	Embeddings engines deployed in your Azure OpenAI Resource
redis_address	api	URL for Redis Stack: "api" for docker composer
redis_password	redis-stack-password	OPTIONAL - Password for your Redis Stack
REDIS_ARGS	--requirepass redis-stack-password	OPTIONAL - Password for your Redis Stack
api_base	https://YOUR_AZURE_OPENAI_RESOURCE.openai.azure.com/	Your Azure OpenAI Resource name. Get it in the Azure Portal
api_key	'YOUR_AZURE_OPENAI_API_KEY'	Your Azure OpenAI Api Key. Get it in the Azure Portal
Blob_Account_Name	'YOUR_AZURE_BLOB_STORAGE_ACCOUNT_NAME'	Get it in the Azure Portal
Blob_Account_Key	'YOUR_AZURE_BLOB_STORAGE_ACCOUNT_KEY'	Get it in the Azure Portal
Blob_Container_Name	'YOUR_AZURE_BLOB_STORAGE_CONTAINER_NAME'	Get it in the Azure Portal
Form_Recognizer_Endpoint	'YOUR_AZURE_FORM_RECOGNIZER_ENDPOINT'	Get it in the Azure Portal
Form_Recognizer_Key	'YOUR_AZURE_FORM_RECOGNIZER_KEY'	Get it in the Azure Portal

How to execute the WebApp and a Redis Stack with docker

• Clone the repo
• In the main folder execute docker compose

    docker compose up

It will spin 2 docker containers:

• WebApp
• Redis Stack for embeddings storing

How to execute the WebApp locally (with an available Redis Stack)

• Create a python env

    python -m venv .venv

• Activate python env

    .venv\Scripts\activate

• Install PIP Requirements

    pip install -r code\requirements.txt

• Set env variable as described in the section above
• Change CWD

    cd code

• Run Streamlit

    streamlit run OpenAI_Queries.py

Hot to execute the WebApp with Docker (with an available Redis Stack)

OPTION 1:

• Run a prebuilt docker image

    docker run -e .env -p 80:80 fruocco/oai-embeddings:latest

OPTION 2:

• Build a docker image

    docker build . -t your_docker_registry/your_docker_image:your_tag

• Run the image

    docker run -e .env -p 80:80 your_docker_registry/your_docker_image:your_tag