workshop-azure-openai

This repository is for a workshop using Azure OpenAI Service.

Prerequisites

To run all the projects in this repository, you need the followings.

• Python 3.10+
• Azure OpenAI Service
• Azure Cosmos DB
• Azure AI Search
• Azure AI Document Intelligence

Here are the preferred tools for development.

• Poetry
• GNU Make

Projects

Project	Description	Image
1_call_azure_openai_chat	Call Azure OpenAI Service API from Python	No Image
2_streamlit_chat	Create an Azure OpenAI Chat app using Streamlit
3_call_azure_cosmos_db	Call Azure Cosmos DB from Python	No Image
4_streamlit_chat_history	Add feature to store chat history using Azure Cosmos DB
5_streamlit_query_chat_history	Search Chat History
6_call_azure_ai_search	Call Azure AI Search from Python	No Image
7_streamlit_chat_rag	Add RAG feature to Streamlit chat app
8_streamlit_azure_openai_batch	Call Azure OpenAI Batch API with Streamlit
9_streamlit_azure_document_intelligence	Call Azure AI Document Intelligence API with Streamlit
10_streamlit_batch_transcription	Call Batch Transcription API with Streamlit
11_promptflow	Get started with Prompt flow	No Image
12_langgraph_agent	Create agents with LangGraph	No Image
99_streamlit_examples	Code samples for Streamlit

How to run

Referring to the .env.template file, create a .env file in the same directory and set the required credentials.

Local environment

# Create a virtual environment
$ python -m venv .venv

# Activate the virtual environment
$ source .venv/bin/activate

# Install dependencies
$ pip install -r requirements.txt

# Run the script (e.g. run 2_streamlit_chat)
$ python -m streamlit run apps/2_streamlit_chat/main.py

Docker container

# Set Docker image name
## GitHub Container Registry
$ IMAGE=ghcr.io/ks6088ts-labs/workshop-azure-openai:latest
## Docker Hub
$ IMAGE=ks6088ts/workshop-azure-openai:latest

# run 2_streamlit_chat
$ docker run --rm \
    -p 8501:8501 \
    -v ${PWD}/.env:/app/.env \
    ${IMAGE} \
    python -m streamlit run ./apps/2_streamlit_chat/main.py