Leading deep learning educator. Tutorials and courses on #DeepLearning, #ComputerVision, #LLMs, #GenAI, #OpenCV, #Keras, #TensorFlow, #PyTorch, and many more.

USA
Joined January 2014
PyImageSearch retweeted
This chapter took us (/w @ritwik_raha) more than two weeks to build. We do not expect you to digest everything at one go, but once you are through, you will be a different person altogether.
We try and join all of the dots in the fifth chapter. This is the most comprehensive chapter in the entire course. Once you are here, take regular breaks and pace yourself properly!
2
5
1,328
PyImageSearch retweeted
Evidence Lower Bound Why do we need it? Where does it come from? Why is it important for diffusion models?
Chapter 4 would be about Latent Variable Modeling. We uncover a very frequently applied concept called "ELBO" here. Also for people enquiring, we will announce our release dates very soon.
1
10
1,224
PyImageSearch retweeted
Nicely written by @ariG23498 and @ritwik_raha!
The third chapter is all about forming an intuition for Generative Modeling. After understanding the idea we then talk about the path to maximum log likelihood. Our focus in the book would be on MLE as the tool to build a generative model.
2
5
1,250
PyImageSearch retweeted
It isn’t about understanding every nuance of the math. Instead focus on building an intuition for using the right tools “Can you hear the music?”
The third chapter is all about forming an intuition for Generative Modeling. After understanding the idea we then talk about the path to maximum log likelihood. Our focus in the book would be on MLE as the tool to build a generative model.
1
3
9
1,697
PyImageSearch retweeted
The dataset that you have has a distribution of its own. How do you mimic that distribution? Why need you need to mimic the distribution? We answer these questions in chapter three.
The third chapter is all about forming an intuition for Generative Modeling. After understanding the idea we then talk about the path to maximum log likelihood. Our focus in the book would be on MLE as the tool to build a generative model.
2
7
1,475
PyImageSearch retweeted
We promise to teach Diffusion from the ground up just like we did with Attention, JAX, NeRF….
All of diffusion rests on the idea of understanding probability distributions and sampling from them. This is often overlooked in papers and such resources, as a known pre-requisite. We begin our second chapter will a primer on various probability distributions.
1
2
18
1,836
PyImageSearch retweeted
Just like we promised—no pre-requisites required. We guide you through everything from the ground up. My favorite part? The 'Notations' section, which really enhances the readability of the entire chapter. 😇
In the first chapter, we focus on Random Variables. We start with an intuition and then branch off into discrete and continuous Random Variables. RVs are foundational to probability theory, and in turn, to understanding diffusion models.
2
14
1,291
Looking forward to it!
1
2
4
763
PyImageSearch retweeted
Waiting for the banger 🫡
2
3
706
PyImageSearch retweeted
Replying to @ritwik_raha
@ritwik_raha and I have dedicated a lot of time to designing this course. Our goal isn’t to teach the latest SoTA model, but to help you understand and read papers on diffusion models. The only prerequisite? An open mind.
We're lowering the entry bar into the world of diffusion with a truly from-scratch course—no prerequisites needed! Not another "you should know KL Divergence"!
2
2
13
1,174
PyImageSearch retweeted
Just crossed 200,000 free public AI datasets shared by the community on @huggingface! Text, image, video, audio, time-series & many more...
8
16
1
157
10,453
PyImageSearch retweeted
Big things are coming! Stay tuned!
After a year worth of work, we will soon be releasing a course on diffusion. Watch this space for more updates!
2
5
1,115
Replying to @PyImageSearch
Looking forward to it!
1
2
2
1,028
PyImageSearch retweeted
The learning curve while building this course was huge. I will always take pride in this project.
After a year worth of work, we will soon be releasing a course on diffusion. Watch this space for more updates!
6
6
34
2,573
Replying to @DeveloperHarris
IIRC, @PyImageSearch was one of the first to make progress on this years ago. As a side note, his Computer Vision and Machine Learning articles are excellent.
1
3
665