Unlock the creative power of generative AI by learning to build your own multimodal systems from the ground up. In this hands-on course, you’ll master deep generative modeling with PyTorch and the Hugging Face ecosystem, progressing from foundational concepts to advanced applications like text-to-image generation and model personalization. Guided by expert instructor Jonathan Dinu, you’ll gain practical skills in manipulating data, training neural networks, and fine-tuning large pre-trained models—empowering you to design innovative AI systems that understand and generate both text and images.

Übungsprojekt

In this specialization, you will embark on a comprehensive, project-based journey to design, build, and personalize your own generative AI systems. Starting with foundational deep learning concepts and the PyTorch framework, you will progressively develop practical expertise by implementing neural networks, convolutional architectures, and variational autoencoders for image generation. You’ll then advance to working with diffusion models and transformer-based architectures, integrating text and image modalities to create powerful multimodal applications. The capstone project will guide you through fine-tuning a pre-trained text-to-image model, such as stable diffusion, enabling you to generate images with unique styles and subjects. By the end of the course, you will have a portfolio-ready generative AI project that demonstrates your ability to build, evaluate, and personalize cutting-edge AI models.