dlaima's picture
Update README.md
5e565cc verified

A newer version of the Gradio SDK is available: 5.35.0

Upgrade
metadata
title: Template Final Assignment
emoji: πŸ•΅πŸ»β€β™‚οΈ
colorFrom: indigo
colorTo: indigo
sdk: gradio
sdk_version: 5.25.2
app_file: app.py
pinned: false
hf_oauth: true
hf_oauth_expiration_minutes: 480
short_description: Agent for GAIA Evaluation

🧠 Agent for GAIA Evaluation

This project contains a Gemini-powered CodeAgent built with smolagents for use in the GAIA Unit 4 Evaluation of the Hugging Face Course.

πŸ§‘β€πŸ« This app is part of the Hugging Face Courses series and was developed as an educational project to showcase how machine learning can be used in real-world event scenarios. Learn more at Hugging Face Courses.


πŸ—‚ Folder Structure

Template Final Assignment_app/
β”‚
β”œβ”€β”€ app.py                       # Main app with Gradio UI and agent execution logic.
β”œβ”€β”€ gaia_tools                   # Custom tools including RunPythonFileTool and download_server.
β”œβ”€β”€ .env.template                # Environment variables template /Template file showing expected environment variables (e.g., API keys).
β”œβ”€β”€ requirements.txt             # Python dependencies. 
└── README.md                    # Project overview, setup instructions, usage examples.

πŸš€ Features

  • Uses the Gemini 2.0 Flash model via LiteLLMModel

  • Equipped with essential tools:

    • DuckDuckGoSearchTool for quick lookups
    • RunPythonFileTool for executing .py scripts
    • ReverseTextTool for decoding reversed questions
    • download_server for fetching files from URLs
    • Base tools (math, string manipulation, etc.)

πŸ“‹ Evaluation Strategy

The agent reads questions from the GAIA evaluation endpoint, applies reasoning using a system prompt with strict guidelines, and submits answers back for scoring.

πŸ› οΈ Setup

  1. Clone this repository or Space

  2. Set your environment variables:

    GEMINI_API_KEY=your_api_key_here
    SPACE_ID=your_hf_space_id
    
  3. Install dependencies:

    pip install -r requirements.txt
    
  4. Run locally:

    python app.py
    

Or launch directly via Hugging Face Spaces.

πŸ§ͺ Evaluation Flow

  1. Log in to Hugging Face through the UI
  2. Click β€œRun Evaluation & Submit All Answers”
  3. The agent will fetch tasks, solve them, and submit results

🀝 Contributing

We welcome contributions to improve GalaGuide!

  1. Fork the repository
  2. Create a new branch:
    git checkout -b feature-xyz
    

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference