Comments (10)
@pixeli99 Thanks for your work. I have a similar question, why you chose the
rand_cosine_interpolated
noise scheduler instead of the one mentioned in theEDM (Karras et al.)
paper. The one highlighted in the following image. Correct me if my understanding is wrong!(Edit: And what does the variables
noise_d_low
andnoise_d_high
correspond to?)Thanks in advance.
Hello, as @Muon2 mentioned, this is the stratege of simple diffusion. You can look at section 3.1 of the paper for more detailed information (since I don't fully understand it either, it might be more reliable to read the original paper directly😢).
from svd_xtend.
@pixeli99 Thanks for your work.
I have a similar question, why you chose the rand_cosine_interpolated
noise scheduler instead of the one mentioned in the EDM (Karras et al.)
paper. The one highlighted in the following image. Correct me if my understanding is wrong!
(Edit: And what does the variables noise_d_low
and noise_d_high
correspond to?)
Thanks in advance.
from svd_xtend.
Thank you very much for raising this question,
This was due to an oversight on my part, as I originally thought that sigma in the code followed a simple log-normal distribution, but I uploaded an incorrect version;
The use of simple diffusion was merely an immature attempt of mine, because my intent was to try mixing videos of different resolutions for training. I wanted to use the strategy of simple diffusion to apply noise of varying distribution (in fact, I am not sure if my understanding is correct,). So, I am actually very eager to ask everyone for their understanding of p_train(σ).
from svd_xtend.
Regarding the second question, it is solely because I have been lazy that I will complete this section of the code.
from svd_xtend.
@pixeli99 Thanks for your work. I have a similar question, why you chose the
rand_cosine_interpolated
noise scheduler instead of the one mentioned in theEDM (Karras et al.)
paper. The one highlighted in the following image. Correct me if my understanding is wrong!
(Edit: And what does the variablesnoise_d_low
andnoise_d_high
correspond to?)
Thanks in advance.
Hello, as @Muon2 mentioned, this is the stratege of simple diffusion. You can look at section 3.1 of the paper for more detailed information (since I don't fully understand it either, it might be more reliable to read the original paper directly😢).
I see. I'll look into the Simple Diffusion paper.
BTW, I am curious have you tried other noise scheduling techniques besides the one mentioned in the Simple Diffusion paper?
from svd_xtend.
@pixeli99 Thanks for your work. I have a similar question, why you chose the
rand_cosine_interpolated
noise scheduler instead of the one mentioned in theEDM (Karras et al.)
paper. The one highlighted in the following image. Correct me if my understanding is wrong!
(Edit: And what does the variablesnoise_d_low
andnoise_d_high
correspond to?)
Thanks in advance.
Hello, as @Muon2 mentioned, this is the stratege of simple diffusion. You can look at section 3.1 of the paper for more detailed information (since I don't fully understand it either, it might be more reliable to read the original paper directly😢).
I think you may be wrong about the simplediffusion sigma sampling. If you are using simplediffusion , you should probably change the sampling scheduler as well, since the Euler sampler is using original EDM timesteps formula instead of the one in the simplediffusion .
from svd_xtend.
I understand what you're saying, but I think that different sigma distributions correspond to different diffusion paths. In theory, would it be possible to use the same sampler for sampling? I suspect there might be a flaw in my understanding, but I'm not sure where I've gone wrong. When we're training, is our definition of the timestep the same, as in 0.25ln(σ)
?
from svd_xtend.
@m-muaz I haven't tried it yet, but if I make any progress, I will update here.
from svd_xtend.
I understand what you're saying, but I think that different sigma distributions correspond to different diffusion paths. In theory, would it be possible to use the same sampler for sampling? I suspect there might be a flaw in my understanding, but I'm not sure where I've gone wrong. When we're training, is our definition of the timestep the same, as in
0.25ln(σ)
?
Sigma distributions would NOT affect diffusion paths actually. The key in simplediffusion is the change of alpha and beta scheduler, just changing the training sigma distributions would not help in my opinion.
from svd_xtend.
I roughly understand what you mean, I might still need to read carefully to grasp the principle here, thank you very much for your reply.
from svd_xtend.
Related Issues (20)
- [CUDA out of memory] training in 1024 × 576 resolution in the A100 80G HOT 4
- Question:How to use the trained modules to generate a video?
- Suggestions on loading video data HOT 1
- 多卡训练报错 HOT 2
- Thank you, fine_tuned weights
- losses barely dropped with BDDx datasets??
- Error when I start the training HOT 2
- Add a short inference script loading from checkpoint? HOT 3
- How can I use the trained module
- How to support batch_size > 1? HOT 1
- 请问下BF16是怎么执行的,我执行有报错,帮看下 HOT 1
- pretrained_model_name_or_path报错 HOT 2
- RuntimeError: Sizes of tensors must match except in dimension 1. Expected size 6 but got size 5 for tensor number 1 in the list. HOT 5
- Question about classifier guidance for image in training code HOT 3
- Exception: Could not find the transformer layer class to wrap in the model.
- after training on 512x512, the video not move always, why? HOT 2
- Training on multi GPUs with different memory size
- Provide requirements.txt by running `pip3 freeze > requirements.txt` from a working development environment HOT 1
- 36gb minimum GPU memory is required using batch size 1 and fp16 mixed precision training? HOT 9
- amazing work!will the part2's code be released?
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from svd_xtend.