Skip to content

Latest commit

 

History

History
127 lines (84 loc) · 4.49 KB

README.md

File metadata and controls

127 lines (84 loc) · 4.49 KB

GenAI Voice-Interactive Data Analytics App

This application provides a seamless voice interaction experience with a large language model (LLM) to analyze data from uploaded CSV files. By integrating multiple components, it offers an interactive and intuitive solution for deriving insights from structured data through voice commands.

Key Features

  • LLM-Powered Analysis
    OpenAI serves as the core language model, processing inputs and providing insights based on extensive language understanding and contextual knowledge.

  • Speech-to-Text Conversion
    Utilizing OpenAI’s Whisper model, the app converts spoken input into text, enabling a smooth voice-driven interaction with the LLM.

  • User-Friendly Interface
    Built on the Streamlit framework, the app provides an intuitive frontend for users to interact with, whether they’re uploading CSV files, issuing voice commands, or viewing analytic results.

  • Data Insight Generation
    Users can upload CSV files and receive analytical summaries and insights powered by the LLM, turning complex data into easy-to-understand insights.

Heart of the GenAI Framework: Swarm

This project is powered by the Swarm framework, a toolkit designed for building ergonomic interfaces for multi-agent systems, making agent coordination and execution lightweight, highly controllable, and easily testable. Swarm enables the development of complex, scalable solutions by using primitive abstractions, such as:

  • Agents
    Each agent encompasses instructions and tools, which makes it possible to represent intricate dynamics between tools and networks of agents. This approach provides a lightweight, scalable, and customizable framework suited for managing large networks of independent capabilities and instructions that can't be condensed into a single prompt.

Customization Highlights

Our customized Swarm setup includes:

  1. Single-File Simplicity
    The core functionality is encapsulated in a single file with less than 390 lines of code, ensuring simplicity and ease of use.

  2. Support for Pandas DataFrames
    Integrates Pandas data frames as context variables, making data handling in analytics straightforward.

  3. Support for Base64 Encoding
    Enables passing any file as a context variable using Base64 encoding, providing flexibility for file management.

  4. Compatibility with Multiple LLMs
    Works with any LLM model compatible with the OpenAI API, including Ollama, Mistral, llama3.2, and more, allowing flexibility in model choice.

Technology Stack and Features

- ✨ [Openai](https://openai.com/) for GenAI services.
- 🐝 [OpenAI Swarm](https://www.langchain.com/) for LLM infrastructure.
- 🦜 [Streamlit](https://streamlit.io/) for the frontend.
- 🧰 [Pandas](https://pandas.pydata.org/) for data analysis and manipulation.
- 🔍 [Pydantic](https://docs.pydantic.dev) for the data validation.
- 🐋 [Docker](https://www.docker.com) for development and production.

Quick start

Run the app in containers

  • Clone the repo and navigate to the root folder.

  • To run the app using Docker, make sure you've got DockerCLI installed on your system. From the project's root directory, run:

    docker build -t z-ask-data .
docker run -d -p 8501:8501 --name ask-data z-ask-data
docker ps

Or, run the app locally

If you want to run the app locally, without using Docker, then:

  • Install Python, version 3.12

  • Clone the repo and navigate to the root folder.

  • Create a virtual environment. Run:

    python3.12 -m venv .venv

  • Activate the environment. Run:

    # Windows
./venv/scripts/activate

# macOS or Linux
source ./venv/bin/activate

  • Install the dependencies. Run:

    pip install -r requirements.txt

  • Start the app. Run:

    streamlit run app.py

  • Go to the following link on your browser:

    http://localhost:8501

Linter and Code formatter

Ruff is extremely fast Python linter and code formatter, written in Rust.

  • Install :

    # On macOS and Linux.
curl -LsSf https://astral.sh/ruff/install.sh | sh 

# On Windows.
powershell -c "irm https://astral.sh/ruff/install.ps1 | iex"

  • Once installed, you can run Ruff from the command line:

    ruff check   # Lint all files in the current directory.
ruff format  # Format all files in the current directory.