Applied AI using Python - Diffusion

Class Description

This course delves into the exciting world of generative models, focusing on the state-of-the-art diffusion models. Students will embark on a journey to understand the theory behind these models and implement a diffusion model from scratch. Through a blend of theoretical concepts and hands-on coding exercises, this course will empower students to create, train, and fine-tune their own diffusion models. By the end of the class, students will have a deep understanding of the mechanics of diffusion models and how they can be applied to various AI tasks. Prerequisites: This boot camp-style class is designed for students in grades 8-12 with beginner to intermediate-level computer skills. Students must be familiar with the fundamentals of Python programming language, basic machine learning concepts and familiarity with frameworks such as TensorFlow or PyTorch, knowledge of neural networks, and linear algebra. Prior experience with generative adversarial networks (GANs) or variational autoencoders (VAEs) is beneficial but not required. What to bring to class: Students are required to bring their own Mac (Mac OS 11+) or Windows-based PC (Windows 10+) with at least 4GB RAM. We will provide students with access to cloud-based GPU hours or in classroom GPU servers for developing practical exercises and projects. Skills you’ll learn: * Theoretical Foundations of Diffusion Models: Understanding the mathematical principles and algorithms underlying diffusion models * Implementation: Hands-on experience in coding a diffusion model from scratch using Python and deep learning libraries * Training and Optimization: Techniques for training diffusion models, including data preparation, hyperparameter tuning, and optimization strategies * Model Evaluation: Methods for evaluating the performance and quality of diffusion models * Applications of Diffusion Models: Insights into real-world applications and use cases of diffusion models in AI * Problem-Solving and Debugging: Developing skills to troubleshoot and debug model implementation issues * Collaboration and Project Management: Working in teams to manage and complete projects, enhancing collaborative and project management skills in a technical setting Cost: $400/month (August - November) Where: Ohlone College, 43600 Mission Blvd, Fremont, CA 94539 Dates: Saturdays, Aug 10 - Nov 23, from 10am - 11:45am (16 classes)