Skip to content

This repository is a comprehensive resource for mastering generative AI, featuring in-depth notes and exciting projects. The goal is to stay updated with the latest advancements in generative AI, and explore applications in image & video generation, creative content creation. Explore the limitless possibilities of generative AI today!

License

Notifications You must be signed in to change notification settings

AMfeta99/Generative_AI

Repository files navigation

Generative AI

Dive into the world of generative AI! Packed with in-depth notes and thrilling projects, this repository is your go-to guide for mastering the latest in generative AI. Stay ahead of the curve with cutting-edge advancements and unleash your creativity in areas like content generation, natural language processing, and synthetic data creation. Explore the limitless possibilities of generative AI today!

image

Index:

Topics_covered

  • Comprehensive Overview of Generative Models
  • Evaluation Metrics
  • Deep Dive into Variational Autoencoders (VAEs) and Generative Adversarial Networks (GANs)
  • Exploration of Advanced GAN Variants like StyleGAN and CycleGAN
  • Diffusion Models and their Advantages Over GANs
  • Implementation Using Hugging Face Diffusers
  • State-of-the-Art Fine-Tuning, Guidance, and Conditioning Methods
  • Specialized Applications with Stable Diffusion: Inpainting, Super-Resolution, and More
  • Cutting-Edge Techniques for Faster Sampling, Enhanced Training, etc

Find my notes: Link

Courses

How Diffusion Models Works (DeepLearning.AI)

  • Create your own diffusion model from scratch.
  • Diffusion process and the models driving it.
  • Pre-built models and APIs.
  • Sampling (DDPM to DDIM), training diffusion models, building neural networks for noise prediction (Unet).
  • Adding context for personalized image generation.

Course Certificate: Link ; More Info.

Diffusion Models Course (by Hugging Face)

  • Explore different types of diffusion models, compare with GANs, and understand their strengths and real-world applications.
  • Implement diffusion models using popular frameworks (Diffusers).
  • Building from scratch versus using established methods like DDPM.
  • SOTA techniques for fine-tuning diffusion models, incorporating guidance and conditioning to enhance control and accuracy.
  • Stable Diffusion, focusing on latent diffusion, text conditioning, and tasks (super-resolution, inpainting, and depth-to-image).
  • Explore improvements in training, like faster sampling via distillation.
  • Methods for video and audio generation.

Course: More Info

Disclaimer

Copyright of all materials in thoses courses belongs to HuggingFace, DeepLearning.AI and can only be used or distributed for educational purpose. You may not use or distribute them for commercial purposes.

Projects:

Here are links to my Generative AI projects:

  • "I see your true colors shining.." The goal is to give literal sense to this song by color B/W images. Feels like magic, right? Let's relive memories, One Color at a Time!

Acknowledgements

About

This repository is a comprehensive resource for mastering generative AI, featuring in-depth notes and exciting projects. The goal is to stay updated with the latest advancements in generative AI, and explore applications in image & video generation, creative content creation. Explore the limitless possibilities of generative AI today!

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published