stylegan2-flax-tpu
denoising-diffusion-pytorch
stylegan2-flax-tpu | denoising-diffusion-pytorch | |
---|---|---|
10 | 11 | |
130 | 7,032 | |
0.8% | - | |
0.0 | 8.5 | |
over 1 year ago | 22 days ago | |
Python | Python | |
- | MIT License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
stylegan2-flax-tpu
-
AI Real-Time Human Full-Body Photo Generator
I’ll take this opportunity to mention our research scaling StyleGAN 2 to larger datasets (using LAION) on food images, leveraging free TPU compute through the TRC program.
We trained for 36 days on a v4-8 on 558k images.
https://nyx-ai.github.io/stylegan2-flax-tpu/
We were hopeful GANs would beat diffusion models when trained on specific domains. But we’ve now switched to Stable Diffusion and Dreambooth training which has proven more much efficient for this purpose.
I still have hopes for GANs! I miss their insane inference speed.
- This Food Does Not Exist
-
[P] This Food Does Not Exist, Updated
https://nyx-ai.github.io/stylegan2-flax-tpu
-
Show HN: Food Does Not Exist, Updated
https://nyx-ai.github.io/stylegan2-flax-tpu
This is an incremental update on the work we shared 3 months ago: https://news.ycombinator.com/item?id=32167704
We keep scaling up StyleGAN2 training using more data, larger models, and more compute. In this release we open-source a 5-class model able to generate images of burgers, cheesecakes, cocktails, cookies, and sushis at resolution 512x512.
This research is part of the technology underlying our AI-generated photography platform Nyx.gallery that we shared here 2 weeks ago: https://news.ycombinator.com/item?id=33179730
It is also part of an academic research into data augmentation using synthetic methods in partnership with the Food & You project (https://www.foodandyou.org/).
-
[P] Towards photorealistic AI images
This is a continuation of our work on "This Food Does Not Exist" (Reddit discussion, Github, checkpoints).
-
Show HN: AI-Generated Photography
Models for the 256px food images were previously released here:
https://github.com/nyx-ai/stylegan2-flax-tpu
-
Food that doesn't exist (StyleGAN2)
This is the continuation of our work on "This Food Does Not Exist", using old-school StyleGAN2 with more data and larger models to generate bigger images (now 512x512 instead of 256x256).
-
Asuka neural net image samples (from NovelAI's in-progress tag-to-image SD model)
I believe you had seen it before on HN: https://nyx-ai.github.io/stylegan2-flax-tpu/
-
[P] This Food Does Not Exist
📘 Repo: https://github.com/nyx-ai/stylegan2-flax-tpu
denoising-diffusion-pytorch
- Commits · lucidrains/denoising-diffusion-pytorch
-
Help using torchaudio and spectrograms for diffusion
I’m trying to train a diffusion model using this code (https://github.com/lucidrains/denoising-diffusion-pytorch). My idea is to take a short audio segment, transform it into a spectrogram and train the model on these images then have it generate spectrograms then go back to audio. However the model requires square images. I cannot for the life of me figure out how to make a square spectrogram. Also is a regular spectrogram or a mel spectrogram better for this application?
-
Implementation of Google's MusicLM in PyTorch
Generally it's without weights, but MusicLM is also a WIP more mature implementations have descriptions on how to train them and follow ups on small scale/crowd-sourced experiments & research[1].
[1]: https://github.com/lucidrains/denoising-diffusion-pytorch
-
[D] Time Embedding in Diffusion Model
[1] https://colab.research.google.com/drive/1sjy9odlSSy0RBVgMTgP7s99NXsqglsUL?usp=sharing#scrollTo=KOYPSxPf_LL7 [2] https://github.com/lucidrains/denoising-diffusion-pytorch/blob/main/denoising_diffusion_pytorch/denoising_diffusion_pytorch.py
-
[D] Can a Diffusion Model be trained with an NVIDIA TITAN X?
Sure. I am using: https://github.com/lucidrains/denoising-diffusion-pytorch
-
[D] Resources to learn and fully understand Diffusion Model Codes
Lucidrains GitHub is always my go to repo for understandable paper implementations https://github.com/lucidrains/denoising-diffusion-pytorch
-
Diffusion model generated exactly the same image as the training image
Thanks for the reply. Is there any suggestion if I wanted to train a model to generate half cat and half butterfly images what I should do? I git cloned the code from https://github.com/lucidrains/denoising-diffusion-pytorch and trained from scratch.
-
[D] Best diffusion model archetype to train?
DDIM/DDPM are the same model to train, they only differ at inference time. To start I would recommend building from lucidrains' MIT licenced version (https://github.com/lucidrains/denoising-diffusion-pytorch). Just play around with the models until you gain an intuition.
-
We just release a complete open-source solution for accelerating Stable Diffusion pretraining and fine-tuning!
Our codebase for the diffusion models builds heavily on OpenAI's ADM codebase , lucidrains, Stable Diffusion, Lightning and Hugging Face. Thanks for open-sourcing!
-
[D] Introduction to Diffusion Models
Once you understand these papers you can begin to understand Palette, and from there I would start with an open-source diffusion implementation like this one and then modify it to suit your needs!
What are some alternatives?
stylegan2-pytorch - Simplest working implementation of Stylegan2, state of the art generative adversarial network, in Pytorch. Enabling everyone to experience disentanglement
ALAE - [CVPR2020] Adversarial Latent Autoencoders
autoregressive - :kiwi_fruit: Autoregressive Models in PyTorch.
tpu-starter - Everything you want to know about Google Cloud TPU
RAVE - Official implementation of the RAVE model: a Realtime Audio Variational autoEncoder
Awesome-Diffusion-Models - A collection of resources and papers on Diffusion Models
pytorch-lightning - Pretrain, finetune and deploy AI models on multiple GPUs, TPUs with zero code changes.
molecule-generation - Implementation of MoLeR: a generative model of molecular graphs which supports scaffold-constrained generation
ColossalAI - Making large AI models cheaper, faster and more accessible
spaCy - 💫 Industrial-strength Natural Language Processing (NLP) in Python