Skillshare
SubscriptionAdobe Firefly: Learn the AI Features of Adobe Creative Apps like Photoshop
Course3.2
Skillshare membership
View course →Coding Stable Diffusion from scratch in PyTorch
No ratings yetBEGINNEREnglish
YouTubeYouTubeMark your own learning status. Opening a course link does not establish enrollment.
Loading saved status…
The course codes Stable Diffusion from scratch in PyTorch and explains the mathematics as it goes. The chapters follow the components in order, from the forward and reverse process and the ELBO loss through the VAE, CLIP, the UNet and a DDPM scheduler.
The description points to a Transformer explainer as the prerequisite, which is worth watching first. The coding chapters are long, so plan this as a series of sessions rather than a single evening.
Full coding of Stable Diffusion from scratch, with full explanation, including explanation of the mathematics. Visual explanation of text-to-image, image-to-image, inpainting
Repository with PDF slides: https://github.com/hkproj/pytorch-stable-diffusion
Prerequisites:
1) Transformer explained: https://www.youtube.com/watch?v=bCz4OMemCcA
Chapters
00:00:00 - Introduction
00:04:30 - What is Stable Diffusion?
00:05:40 - Generative Models
00:12:07 - Forward and Reverse Process
00:17:44 - ELBO and Loss
00:20:30 - Generating New Data
00:22:20 - Classifier-Free Guidance
00:31:00 - CLIP
00:33:20 - Variational Auto Encoder
00:37:26 - Text to Image
00:39:54 - Image to Image
00:41:40 - Inpainting
00:44:30 - Coding the VAE
01:54:50 - Coding CLIP
02:09:10 - Coding the Unet
03:04:40 - Coding the Pipeline
03:53:00 - Coding the Scheduler (DDPM)
04:38:00 - Coding the Inference code
Advertisement
No Korshub reviews yet.
Adobe Firefly: Learn the AI Features of Adobe Creative Apps like Photoshop
