Lesson 9: Deep Learning Foundations to Stable Diffusion

[@rajahx]

This is beautifully explained Jeremy! From real basics to some of the most complicated state of the art models we have today. Bravo.

[@TTTrouble]

Wow this is such a treasure to have freely available and I am so thankful that you put this out for the community. Many many thanks good sir, your work towards educating the masses about AI and Machine Learning is so very much appreciated. 🎉❤

[@flavioferlin3127]

Such a brilliant, humble, altruistic, knowledgeable person is a Godsend. Thank you Sir! Every time you watch this you can’t help but appreciate his way of teaching and being inspired to follow the outstanding work.

[@numannebuni]

I absolutely love the style in which this is explained. Thank you very much!

[@howardjeremyp]

Glad you like it!

[@mamotivated]

Liberating the world with this quality of education

[@gilbertobatres-estrada5119]

I am so glad you took your time to correct the math mistake! Great work! And thank you for your mission of teaching us new findings in AI and deep learning 🙏

[@MuhammadJaalouk]

Thank you so much for this insightful video. The lecture breaks down complex ideas into segments that are very easy to comprehend.

[@AIBites]

This is a nicely thought-through course. Amazing Jeremy! :)

[@chyldstudios]

Wonderful, I was waiting for these series of videos. Bravo!

[@ItzGanked]

I thought I was going to have to wait until next year, thank you for making this content accessible

[@asheeshmathur]

Outstanding, the best description so far. God Bless Jeremy. Excellent service to curious souls.

[@kartikpodugu]

🙏🙏🙏 Amazing information. I knew bits and pieces, now I know the entire picture.

[@cybermollusk]

You might want to put this series into a playlist. I see you have playlists for all your other courses.

[@marko.p.radojcic]

I am getting KZbin premium just! to download this series. Thank you!

[@sushilkhadka8069]

Excellent intiution. You're doing the huge service to humanity

[@peregudovoleg]

At 1:13:20 aren't we supposed to add derivatives to pixel values since we are maximizing P? Unless, since P is binary and it looks like a classification problem, we are going to get negative logits, then deducting seems ok (not touching the sign). Great course!

[@sotasearcher]

28:36 - I'm here in February '24, where they are good enough to do it in 1 go with SDXL-Turbo / ADD (Nov '23) :)

[@danielhabibio]

2024, the fundamentals are still sound. Thank you Jeremy and team!

[@akheel_khan]

Undoubtedly an accessible and insightful guide

[@208muppallahindu5]

Thank you , Jeremy Howard for teaching me concepts of diffusion.

[@ricardocalleja]

Awesome material! Thank you very much for sharing

[@howardjeremyp]

My pleasure!

[@ghpkishore]

That math correction was very essential to me. Coming from a mechanical background, I knew something was off, but then thought I didn't know enough about DL to figure out what it is, and that I was on the wrong. With the math correction, it clicked, and was something I knew all along.Thanks.

[@sushilkhadka8069]

at 1:56:50 I'm having hard time understanding the cost function. I think we need to maximise ( Green Summation - Red Summation ) , for that reason we can't call it a cost function because cost functions are usually minimised. Please correct me If I'm wrong

[@Omunamantech]

Marvellous, Beautiful, Loved It :D

[@ayashiyumi]

Ottimo video. Continua a pubblicare altre cose del genere.

[@johngrabner]

Very informative video. Thank you for taking the time to produce.

[@SadAnecdote]

Thanks for the early release

[@rashulsel]

Amazing video and really easy to follow up with the topics. Its neat how different research is coming together to build something more efficient and promising. So future of AI is how models fit together?

[@HoomanRafraf]

Thanks for the awesome elaboration. There's one thing that was not clear to me in the process and I wasn't quite able to find the answer by googling: Why do we need CLIP and can't use a regular text encoder? As far as I can see, in this architecture what matters for the role of the text encoder is to give similar embeddings for similar contexts, e.g. "a beautiful swan" and "such a lovely swan" should have similar embeddings. But why should this embedding also be similar to the embeddings for images of the swan?

[@SubhadityaMukherjee]

YAY its hereeee. My excitement!!

[@mariuswuyo8742]

Very an excellent course, I would like to ask a question that the noise N(0,0.1) is added equally to each pixel or to the whole image at 1:21:51? These two are equivalent？

[@user-wf3bp5zu3u]

Different per pixel! You’re drawing a vector of random noise samples then reshaping it into an image, so you get many values but all from a dist with low variance. The python random numbers let you sample in the shape of an image directly, so you don’t need to manually reshape. But that’s just for convenience

[@atNguyen-gt6nd]

Thank you so much for your lectures.

[@kirak]

Wow this helped me a lot. Thank you!

[@edmondj.]

I love you, its so clear as usual, i owed you embeddings, now i owe you diffusion too.

[@edmondj.]

Please open a tipee

[@useless_deno]

Amazingly explained !

[@pranavkulkarni6489]

Thank you for great explanation .. I just wanted to know ans to 'what is U-net?' could not understand where is it used in whole process ? I mean what I could not get is what is the difference between VAE (Autoencoder) and an Unet

[@tildebyte]

During *training*, you pass an actual image into the VAE ENcoder (to reduce the amount of data you have to deal with), which then passes the latent it produces on to the UNet, which does the learning involving noising/denoising the latent. During *inference* ("generating"), the UNet (after a lot of other stuff happens :D) passes out a denoised latent to the VAE DEcoder, which then produces an actual image

[@adityagupta-hm2vs]

Why do we predict the noise and not the number as the output itself ? Is it because loss calculation is easier with that ?

[@ramansarabha871]

Thanks a ton! have been waiting.

[@howardjeremyp]

Hope you like it!

[@pankajsinghrawat1056]

since we want to increase the probability of our image being a digit, we should "add" and not "substract" the grad of probability wrt to img. Is this right? or am I missing something ?

[@homataha5626]

Can I ask a video of these model that is used for colonization?

[@andrewimanuel2838]

where can I find the latest recommended cloud computing resource?

[@tinkeringengr]

Thanks -- great lecture!

[@mikhaeldito]

Released already??

[@jonatan01i]

Good thing is that with git we can go back to the state of the code as of (11/20).10.2022

[@ДмитроПавличко-п9д]

Thank you for this lecture

[@tildebyte]

I've been working on/with diffusion models (and before that VQGANs!) for years now, so I'm pretty familiar (from the end-user/theoretical POV, not so much the math/code side heh) with samplers/schedulers - this is the first time I've conceived of them as optimizers, and that seems like a *really* fertile area to research. Have you (or anyone else, for that matter) made any progress in this direction? It's (not too surprisingly) VERY hard to prompt today's search engines to find anything to do with "denoise|diffusion|schedule|sample|optimize" and NOT come up with dozens of either HuggingFace docs pages, or pages w.r.t. Stable DIffusion ROFL

[@edwardhiscoke471]

Already out, wow. Then I'd better push on with part 1!

[@super-eth8478]

THANKS 🙏🏻🙏🏻

[@rubensmau]

Thanks, very clear.

[@kawalier1]

Jerremy, Adam has eps, SGD momentum

[@sotasearcher]

52:12 - The upside down triangle is "nabla", "del" is the squiggly d that goes before each partial derivative. Also, I'm jealous of people who started calculus in high school lol

[@sotasearcher]

Nevermind! Getting to the next section edited in lol

[@sotasearcher]

1:04:12 wait you still mixed them up 😅 At this rate with your following, you're going to speak it into existence though lol. Math notation ultimately is whatever everyone agrees upon, so I could see it being possible.

[@sotasearcher]

1:05:22 - While I'm being nit-picky - Right-side-up triangle is called "delta", and just means change, not necessarily small

[@gustavojuantorena]

👏👏👏

[@mohdil123]

Awesome

[@AndrewRafas]

There is a small (not that important) correction: when you talk about 16384 bytes of latents, it is 16384 numbers, which are 65536 bytes in fact.

[@sambitmukherjee1713]

Each number is 4 bytes because it's a float32 precision?

[@flavioferlin3127]

Wow!

[@susdoge3767]

gold

[@TiagoVello]

UHUUUL ITS OUT

[@JingyingAIEducation]

I was wondering if I could give some suggestions, you spent 20 mins to explain the course materials and different people. Why not start from main fun play first, then later introduce more about the course materials. People will lose interest to listen to 20 mins course materials.

[@yufengchen4944]

Great! I can only see 2019 version of Part 2, look foward to see the new Part 2 course available!

[@yufengchen4944]

Looks like the part 2 2022 webpage is still not public right? or I didn't find the way?

[@michaelnurse9089]

I thought I was going to have to wait until next year, thank you for making this content accessible

[@Beyondarmonia]

Thank you 🙏