Link to the code: github.com/dome272/Diffusion-Models-pytorch

@bao-dai

Жыл бұрын

21:56 The way you starred your own repo makes my day bro 🤣🤣 really appreciate your work, just keep going!!

@outliier

Жыл бұрын

@@bao-dai xd

@leif1075

Жыл бұрын

@@outliier Thanks for sharing but how do you not get bored or tired of doing the same thing for so long and deal with all the math?

@outliier

Жыл бұрын

@@leif1075 I love to do it. I don’t get bored

@ananpinya835

Жыл бұрын

After I saw your next video "Cross Attention | method and math explained", I would like to see ControlNet's openpose in PyTorch Implementation which control posing on image of a dogs. Or if it is too complicate, you may simplify it to control 2 - 3 branches shape of a tree.

Hello, this has become a great video once again. We didn't understand much, but it's still nice to watch. Greetings from home say Mam & Dad. ;-))))

Great, this video is finally out! Awesome coding explanation! 👏

This video is really timely and needed. Thanks for the implementation and keep up the good work!

thank you so much for your detailed explaination of the code. It helped me a lot on my way of learning diffusion model. Wish there are more youtubers like you!

Incredible. Very thorough and clear. Very, very well done.

These implementation videos are marvelous. You really should do more of them. Big fan of your channel!

Congrats, This is a great channel!! hope to see more of these videos in the future.

After my midterm week i wanna study diffusion models with your videos im so exited .thanks a lot for good explanation

most informative and easy to understand video on diffusion models on youtube, Thanks Man

This channel seems to be growing very fast. Thanks for this amazing tutorial.🤩

Thank you for sharing the implementation since authentic resources are rare

great tutorial! looking to seeing more of this! keep it up!

We chose Diffusion Model as part of our course project, and your videos do save much of my time to understand the concepts and have more focus on implementing the main part. I am really grateful for your contribution.

Sincere gratitude for this tutorial, this has really helped me with my project. Please continue with such videos.

This videos is crazy! I don't get tired of recommend it to anyone interesting in diffusion models. I have recently started to research with these type of models and I think your video as huge source of information and guidance in this topic. I find myself recurrently re-watching your video to revise some information. Incredible work, we need more people like you!

@outliier

5 ай бұрын

Thank you so much for the kind words!

Amazing tutorial, very informative and clear, nice work!

I was wating for so long i learnd about condicional difusion models

Dude, you're amazing! Thanks for uploading this!

Very helpful walk-through. Thank you!

The best video for diffusion! Very Clear

Thank you. Best explanation with good DNN models

this is the most underrated channel i've ever seen, amazing explanation !

@outliier

Жыл бұрын

thank you so much!

Thank you so much for this sharing, that was perfect!

Looking forward for some video on Classifier Guidance as well. Thanks.

Very well done! Keep the great content!!

nice demonstration, thanks for sharing

Thanks, this implementation really helped clear things up.

Incredible explanation, thanks a lot!

awesome implementation!

Very nicely explained. Thanks.

The Under rated OG channel

It's definitely cool and helpful! Thanks!!!

Hey! I am start my CompSci Masters program in the Fall, and just wanted to say that I love this video. I've never really had time to sit down and learn PyTorch, so the brevity of this video is greatly appreciated! It gives me a fantastic starting point that I can tinker around with, and I have an idea on how I can apply this in a non-conventional way that I haven't seen much research on... Thanks again!

@outliier

29 күн бұрын

Love to hear that Good luck on your journey!

Fantastic video!

thanks for your amazing efforts!

Awesome video.

Amazing stuff!

Thank you very much, it has solved my urgent need

one CRAZY thing to take from this code (and video) GREEK LETTERS ARE CAN BE USED AS VARIABLE NAME IN PYTHON

Great video!

it is very helpful!! You are a genius.. :) thank you!!

Thank you for sharing!

Hi, @Outlier , thank you for the awesome explanation ! Just one observation, I believe in line 50 of your code (at 19:10) it should be: uncond_predicted_noise = model(x,t,None) 😁

@outliier

Жыл бұрын

good catch thank you. (It's correct in the github code tho :))

Great video!! You make coding seem like playing super mario 😂😂

great walkthrough. but where would i implement dynamic or static thresholding as described in the imagen paper? the static thresholding clips all values larger then 1 but my model regularly outputs numbers as high as 5. but it creates images and loss decreases to 0.016 with SmoothL1Loss.

great video, you got one new subscriber

Could you please explain the paper "High Resolution Image Synthesis With Latent Diffusion Models" and its implementations? Your explanations are exceptionally crystal.

This is my first few days of trying to understand diffusion models. Coding was kinda fun on this one. I will take a break for 1-2 months and study something related like GANs or VAE, or even energy-based models. Then comeback with more general understanding :) Thanks !

@zenchiassassin283

8 ай бұрын

And transformers for the attention mechanisms + positional encoding

@TheAero

8 ай бұрын

I got that snatched in the past 2 months. Gotta learn the math, what is actually a distribution etc.@@zenchiassassin283

This is GOLD

Thank you so much for this amazing video! You mention that changing the original DDPM to a conditional model should be as simple as adding in the condition at some point during training. I was just wondering if you had any experience with using DDPM to denoise images? I was planning on conditioning the model on the input noisy data by concatenating it to yt during training. I am going to try and play around with your github code and see if I can get something to work with denoising. Wish me luck!

Thank you!!

Thank you so much for this amazing video! In mention that the first DDPM paper show no necessary of lower bound formulation, could you tell me the specific place in the paper? thanks!

Super cool

Nice tutorial

Thank you for the video. How can we use diffusion model for inpainting?

This video is priceless.

terrific！

Thank you for the review. So, what is the key to make a step from text description to image? Can you please pinpoint where it is explained?

Thanks alot :)

can you do a text to image in small dataset similar to SD from scratch?

Can you please explain how to use Woodfisher technique to approximate second-order gradients? Thanks

Can you please tell me how much time was need to train this 3000 image for 500 Epoch?

Awesome! How did you type Ɛ in code?

Hi , I want to use a single underwater image dataset what changes do i have to implement on the code?

Great Video, On what Data did you train your model again?

Can you do one for tensorflow too btw very good explaination

Why is the bias off in the initial convolutional block?

How can i increase the img size to 128 pixels square?

having hard time to understand the mathematical and code aspect of diffusion model although i have a good high level understanding...any good resource i can go through? id appreciate it

How can i increase the size of the generated image here?

is anyone find the DDPM Unet architecture figure, I can't find it

Hello, thanks for your a lot contribution ! But a bit confused, At 06:04, just sampling from N(0, 1) totally randomly would not have any "trace" of an image. How come the model infer the image from the totally random noise ?

@outliier

Жыл бұрын

Hey there, that is sort of the "magic" of diffusion models which is hard to grasp your mind around. But since the model is trained to always see noise between 0% and 100% it sees full noise during training for which it is then trained to denoise it. And usually when you provide conditioning to the model such as class labels or text information, the model has more information than just random noise. But still, unconditional training still works.

last self attention layer (64, 64) changes my training type from 5 minutes to hours per epoch, do you know why? training on a single rtx 3060 TI gpu

So the process of adding noise and removing it happens in a loop

hey can we use an image as a condition

` x = torch.randn((n, 3, self.img_size, self.img_size)).to(self.device) predicted_noise = model(x, t) ` in the deffusion class why you create an noise and pass that noise into the model to predict noise ... please explain

best diffusion youtube

Great video, thanks for making it. I started working with diffusion models very recently and I used you implementation as base for my model. I am currently facing a problem that the MSE loss starts very close to 1 and continues like that but varying between 1.0002 and 1.0004, for this reason the model is not training properly. Did you face any issue like this one? I am using the MNIST dataset to train the network, I wanted to first test it with some less complex dataset.

@justinsong3506

Жыл бұрын

I am facing similar problems. I did the experiment on CIFAR10 dataset. The mse loss starts descresing normally but at some points the loss increse to 1 and never descrese again.

Wonderful video! I notice that at 18:50, the equation for the new noise seems to differ from Eq. 6 in the CFG paper, as if the unconditioned and conditioned epsilons are reversed. Can you comment on that?

I think your code bugs when adjust image_size?

Your videos are a blessing. Thank you very much!!! Have you tried using DDIM to accelerate predictions? Or any other idea to decrease the number of steps needed?

@outliier

Жыл бұрын

I have not tried any speedups in any way. But feel free to try it out and tell me / us what works best. In the repo I do linked a fork which implements a couple additions which make the training etc. faster. You can check that out too here: github.com/tcapelle/Diffusion-Models-pytorch

@remmaria

Жыл бұрын

@@outliier Thank you! I will try it for sure.

How do you use this models to generate text to image?

@outliier

Жыл бұрын

You would need to train it on text-image pairs instead of label-image pairs as in the video. And you would need to scale up the model and dataset size to get some nice results

Like your channel, please make more videos

With this training method, wouldn't there be a possibility of some timesteps not being trained in an epoch? wouldn't it be better to shuffle the whole list of timesteps and then sample sequentially with every batch?

The best

Thx Mr gigachad

Hi! Can you please explain why the output is getting two stitched images?

@outliier

3 ай бұрын

What do you mean with two stitched images?

Hey, I am getting an error when i try to use one channel "RuntimeError: Given groups=1, weight of size [64, 1, 3, 3], expected input[4, 3, 64, 64] to have 1 channels, but got 3 channels instead" What can I do?

@outliier

Жыл бұрын

You need to change the input and output channels in the unet code

@outliier Do you think there is a way to run the code with a 3060 GPU on personal desktop? I get the error message: CUDA out of memory.

@MrScorpianwarrior

29 күн бұрын

Random person 6 months later, but you could try decreasing the batch size during training. Your results may not look like what he got in the video though!

Very cool. How would DDIM models be different? Do they use a deterministic denoising sampler?

@outliier

6 ай бұрын

yes indeed

6:57 Why the formula is ... + torch.sqrt(beta) instead of calculated posterior variance like in paper?

@outliier

Жыл бұрын

Which paper are you referring to? In the first paper, you would just set the variance to beta and since you add the std * noise you take the sqrt(beta)

8:38 in the UNet section, how do you decide on the number of channels to set in both input and output to the Down and Up classes. Why just 64,128, etc. ?

@outliier

10 ай бұрын

People just go with powers of 2 usually. And usually you go to more channels in the deeper layers of the network.

@kashishmathukiya8091

10 ай бұрын

@@outliier oh okay got it. Thank you so much for clearing that and for the video! I had seen so many videos / read articles for diffusion but yours were the best and explained every thing which others considered prerequisites!! Separating the paper explanation and implementation was really helpful.

Thank you very much for this very easy-to-understand implementation. I have one question: I don't understand the function def noise_images. Assume that we have img_{0}, img_{1}, ..., img_{T}, which are obtained from adding the noise iteratively. I understand that img{t} is given by the formula "sqrt_alpha_hat * img_{0} + sqrt_one_minus_alpha_hat * Ɛ". However, I don't understand the function "def noise_images(self, x, t)" in [ddpm.py]. It return Ɛ, where Ɛ = torch.randn_like(x). So, this is just a noise signal draw directly from the normal distribution. I suppose this random noise is not related to the input image? It is becasue randn_like() returns a tensor with the same size as input x that is filled with random numbers from a normal distribution with mean 0 and variance 1 In training, the predicted noise is compared to this Ɛ (line 80 in [ddpm.py]). Why we are predicting this random noise? Shouldn't we predict the noise added at time t, i.e. "img_{t} - img_{t-1}"?

@Laszer271

Жыл бұрын

I had the same misconception before. It was actually explained by "AI Coffee Break with Letitia" channel in a video titled "How does Stable Diffusion work? - Latent Diffusion Models EXPLAINED". Basically, the model tries to predict the WHOLE noise added to the image to go from noised image to a fully denoised image in ONE STEP. Because it's a hard task to do, the model does not excel at that so at inference we denoise it iteratively, each time subtracting only a small fraction of the noise predicted by the model. In this way, the model produces much better quality samples. At least that's how I understood it :P

@rikki146

Жыл бұрын

@@Laszer271 While I understand it predicts the "whole noise", this "whole noise" is newly generated and I suppose the ground truth is (img_{t} - img_{0)).. still can't wrap my head around it.

Hi, Thank you for the Video! Can you please explain the test part: n = 4 device = "cpu" model = UNet_conditional(num_classes=4).to(device) ckpt = torch.load(r"C:\Users oueft\Downloads\Diffusion-Models-pytorch-V7\models\DDPM_conditional\ckpt.pt", map_location=torch.device('cpu')) model.load_state_dict(ckpt) diffusion = Diffusion(img_size=64, device=device) y = torch.Tensor([6] * n).long().to(device) x = diffusion.sample(model, n, y) plot_images(x) What is n, and why did the following error come up when I ran it? ddpm_conditional.py", line 81, in sample n = len(labels) TypeError: object of type 'int' has no len()

Thank you...U just made diffusion so easy to understand... I would like to ask ; What changes do I need to make in order to give an image as condition rather than a label as condition. I mean how to load ground Truth from GT repository as label (y).

@outliier

Жыл бұрын

Depends on your task. Could you specify what you want to achieve? Super resolution? Img2Img?

@Naira-ny9zc

Жыл бұрын

@@outliier I want to generate thermal IR images conditioned on their respective RGB images . I know that in order to achieve this task i.e ; Image (RGB) to Image (Thermal IR) translation, I have to concat the input to U-net (which of course is thermal noise image ) with its corresponding RGB (condition image) and give this concatenated output as final input to the unet ; but problem is that I am not able to put this all together in the code (especially concatenating each RGB image (condition) from RGB image folder with its corresponding Thermal noise images so that I can pass the concatenated resultant image as final input to Unet as my aim is to generate RGB conditioned Thermal image using Diffusion.

Great video! How long did it take to train the models?

@outliier

Жыл бұрын

About 3-4 days on an rtx 3090.

thanks for the easiest implementation. could you plz tell us how to find FID and IS score for these images?

@outliier

8 ай бұрын

I think you would just sample 10-50k images from the trained model and then take 10-50k images from the original dataset and then calculate the FID and IS

@muhammadawais2173

8 ай бұрын

@@outliier thanks

You do not use any LR scheduler. Is this intentional? My understanding is that EMA is a functional equivalent of LR scheduler, but then I do not see any comparison between EMA vs e.g. cosine LR scheduler. Can you elaborate more on that?

Great videos on diffusion models, very understandable explanations! For how many hours did you train it? I tried adjusting your conditional model and train with a different dataset, but it seems to take forever :D

@outliier

Жыл бұрын

Yea it took quite long. On the 3090 it trained a couple days (2-4 days I believe)

@maybritt-sch

Жыл бұрын

@@outliier Thanks for the feedback. Ok seems like I didn't do a mistake, but only need more patience!

@outliier

Жыл бұрын

@@maybritt-sch Yea. Let me know how it goes or if you need help