Delve into Xilinx's AI model zoo to harness powerful AI models, guiding you through a progressive learning path. These modules build directly toward this year's Physical AI theme and the GridMind Robotics Competition, where everything you learn here runs live on your own board.
How to Navigate the Trainings
To get the most out of this curriculum, we recommend following the trainings in the order presented. Each training module builds upon the knowledge and skills acquired in the previous ones, ensuring a smooth and effective learning experience.
Training Modules
PYNQ 101 - Intro to PYNQ
Start here to get a solid foundation in PYNQ, understand its key features, and learn how to set up your environment.
PYNQ 201 - MNIST
Learn how to work with the MNIST dataset, implementing and understanding basic image classification.
PYNQ 505 - Prompt Engineering
Learn how to write effective prompts for LLMs, including the free-hint and riddle solvers your GridMind client will use.
PYNQ 301 - Object Detection
Dive into object detection techniques, utilizing pre-trained models to detect objects in images and videos — the core vision skill behind your GridMind client.
PYNQ 504 - Board to Board Communication
Learn the referee wire protocol your board uses to join a match, report moves, and receive turn/score updates in real time.
Breakout 1: Advanced Topics
- PYNQ 502 - Intermediate Python
- Enhance your Python programming skills, essential for effective AI model implementation.
- PYNQ 506 - Genesis Robot Simulation
- See your GridMind client's moves animated by a simulated robot arm.
Breakout 2: Vision Deep Dive
- PYNQ 501 - Computer Vision
- Advanced computer vision techniques, including more complex image processing and analysis.
- PYNQ 503 - Pose Detection
- Learn about pose detection algorithms, enabling you to detect and analyze human poses in real-time.
Prior-Year / Self-Paced Modules
Not part of this year's live schedule, but still available if you want to explore further on your own:
- PYNQ 401 - Crowd Estimation — estimate crowd sizes in images and videos.
- PYNQ 402 - Face Mask Detection — detect face masks using AI models.
- PYNQ 403 - Automotive — vehicle detection and traffic sign recognition.
Getting Started
Setup Your Environment
- If you want to set up your own image, start here: KRIA-SOM, PYNQ Setup, and Bootcamp Image Creation
- If you would like to use a pre-setup image and get started with the trainings, start here: Getting Started with PYNQ Bootcamp
Begin with PYNQ101: Start your learning journey with the introductory module and progressively move through the modules.
Tips for Success
- Take Your Time: Don't rush through the modules. Take your time to understand each concept thoroughly.
- Hands-On Practice: Actively engage with the practical exercises and projects.
- Ask Questions: Utilize the community forums and support channels if you encounter any issues or have questions.
- Review and Reflect: Regularly review your progress and reflect on what you've learned.
Embark on this exciting journey to master the KRIA KV260 with PYNQ, and unlock the full potential of AI and machine learning in your projects!